the value of housing design and layout - the national...

By 2016 an additional three million homes will be required in the UK. Yet last

year an estimated 160,000 new homes were built, far short of the

Government’s target of 225,000.

This report, produced by FPDSavills Research for CABE and ODPM in

association with Design for Homes, seeks to demonstrate a way of measuring

the economic dividends of investing in high quality housing design and layout. It

pilots a methodology that developers, funders, planners and housebuilders can

apply to compare housing sites and schemes.

The report is especially timely given pressure from the Government who rightly

insist that new housing development, especially in the south east of England,

should be aiming for densities of over 30 units per hectare. Investing in the

design of these new homes is imperative if we are to deliver quality

neighbourhoods of tomorrow.

ThomasTelford

the value of housing design and layout



a research project commissioned by CABE and ODPM in association with Design for Homes

produced by FPDSavills Research with Davis Langdon & Everest

ThomasTelford


CABE, the Commission for Architecture and the Built Environment, is the nation’s champion for better places, places which

• work better

• feel better

• are better

We believe that decent homes, work places, shops, schools and hospitals are everyone’s basic right, a right worth fighting for. CABE uses its skills and resources to

campaign for a better quality of life for people and communities across England. We do this through a rich mix of campaigning, researching, instigating and assisting

with technical expertise and opinion.

CABE

Tower Building

11 York Street

London SE1 7NX

Telephone 020 7960 2400

www.cabe.org.uk

ODPM

Eland House

Bressenden Place

London SW1E 5DU

www.odpm.gov.uk

Design for Homes

79 Parkway

London NW1 7PP

www.designforhomes.org

Copies of this report are available on CABE’s website at www.cabe.org.uk

or from:

The Customer Services Department

Thomas Telford Limited, Units I/K

Paddock Wood Distribution Centre

Paddock Wood, Tonbridge

Kent TN12 6UU

Tel: 020 7665 2464

Fax: 020 7665 2245

www.thomastelford.com

ISBN 07277 3208 0

© Commission for Architecture and the Built Environment, 2003

Although every care has been taken in preparing this report with information obtained from sources considered to be reliable, it has not been independently verified

by CABE, its officers, employers, agents and advisers. No representation, warranty or undertaking, express or implied, is made and no responsibility or liability will be

accepted by CABE or its employees, agents and advisers as to the accuracy or completeness of this report or any other written or oral information made available.

Whilst every effort has been made to ensure that the data contained in this report is correct, no responsibility can be taken by the authors or publisher for omissions

or erroneous data provided by a third party or due to information being unavailable or inaccessible during the research period.

The statements and opinions presented here are those of the researchers and not necessarily those of CABE, ODPM or the publisher.

All rights, including translation, reserved. Except as permitted by the Copyright, Designs and Patents Act 1988, no part of this publication may be reproduced, stored

in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying or otherwise, without the prior written permission of the

Publishing Director, Thomas Telford Publishing, Thomas Telford Ltd, 1 Heron Quay, London E14 4JD.

Design by Kneath Associates

Printed in Great Britain by Latimer Trend

ThomasTelford

T H E V A L U E O F H O U S I N G D E S I G N A N D L A Y O U T 3

England will need at least another three million homes by 2016. Yet last year an

estimated 160,000 were built, far short of the Government’s target of 225,000.

National opinion surveys consistently show that a significant section of the house

buying public would never consider purchasing a new house, preferring more

established neighbourhoods and building stock.

We have to accept that to date we have not provided a product that enough people

are willing to buy into. Where demand so grossly outstrips supply, there is little

commercial incentive for housebuilders to offer more innovative and attractive

designs. So consumers are left without a choice.

But we need to take account of the fact that people see value in good design and

that this will attract them to modern housing. Innovative thinking, integration with

existing communities and investment in quality are the things which will persuade

people that they want to live in modern housing.

Well designed housing improves quality of life for residents and the wider

community, helps reduce crime, fosters community cohesion and significantly

reduces whole life costs.

This CABE/ODPM commissioned report seeks to demonstrate a way of measuring


provides a methodology that developers, funders, planners and housebuilders can

use for comparing different housing sites and schemes.

It is especially timely given pressure from the Government who rightly insist that new

housing development, especially in the south-east of England, should be aiming for

densities of over 30 units per hectare.

We do not want to repeat recent mistakes – soulless estates, cramped housing and

inefficient neighbourhood layouts. We want to see houses where people aspire to

live and stay. Investing in the design of these new buildings and, just as importantly,

in the quality of the public space between them is imperative if we are to deliver the


Sir Stuart Lipton

Chairman

CABE

Sir Stuart Lipton

Chairman

CABE

Foreword

F O R E W O R D

T H E V A L U E O F H O U S I N G D E S I G N A N D L A Y O U T

C O N T E N T S

5

Executive summary 7

1.0 Introduction 11

1.1 Context 12

2.0 Addressing the gaps 15

2.1 Introduction 16

2.2 Housebuilding and design: institutional context 16

2.3 Brownfield land development and housing 19

2.4 Housing design quality and ‘value’ 21

2.5 Summary 23

3.0 Methodology 25

3.1 Overview 26

3.2 Selection of case studies 27

3.3 Dealing with externalities 29

3.4 Measuring and quantifying the sites 30

3.5 Measuring costs 31

3.6 Measuring value 31

3.7 Assessing commercial value 33

4.0 Research results and discussion 35

4.1 Commercial value 36

4.2 Scheme characteristics 36

4.3 Costs 41

4.4 Values 42

5.0 Conclusions 45

Appendix 1 Literature review references 49

Appendix 2 Cost analysis by Davis Langdon & Everest 53

A2.1 Introduction 54

A2.2 Objective 54

A2.3 Data collection 54

A2.4 Scope and structure of project cost data 55

A2.5 Analysis of the data 56

A2.6 Results 57

Appendix 3 Characteristics of each site 65

Appendix 4 Step-by-step methodology 83

Appendix 5 Illustrative example of the residual land calculation 85

Acknowledgements 87

Contents

E X E C U T I V E S U M M A R Y

Executive Summary


8

This report attempts to assess the extent to which design changes the value of

residential development. It sets out a methodology whereby different types of design

in housing and housing layout are examined to see how their differences affect the

commercial value of new developments for developers.

This research pilots a methodology that can be used to compare housing on any

site in any location. Developers, funders, planners and other interested parties will

be able to use this methodology on any site to make a like-for-like comparison

between different developments.

The methodology and the results from the small sample are intended to act as a

framework for further, ongoing investigation into the role of good design in housing

(or what is deemed to be good design) and its effects on value.

The focus of this study is to demonstrate what, if any, enhanced value there is for the

developer in building well designed schemes, broadly upholding the urban design

principles in PPG3, compared to building more recent standard examples of housing

layout and design. It is not concerned with whether the capital values of individual

properties may or may not be higher than a comparably sized, standard new home.

Neither does it examine the amount of capital appreciation that individual homes within

the more innovative schemes may enjoy compared to more standard properties.

The research selected eight examples of completed developments and compared

them in four matched pairs. It was intended that four schemes with certain

‘innovative’ design elements would be compared against four of a more ‘standard’

design in the same location. Practical difficulties in obtaining confidential commercial

information from certain developers necessitated creating two hypothetical

comparator schemes sharing the same locations as two of the schemes with

innovative design elements. The remaining two matched pairs both comprised real

schemes. The eight sites were studied in four different locations around the South

East. This had the effect of eliminating the impact of regional economies and the

differences between regional housing markets when making comparisons.

The developments were costed and a market valuation made. A residual value was

then calculated on each scheme and expressed on a per hectare of land basis

(using the same overall measure of land for each example) so that all sites could be

compared to each other on a like-for-like basis. The residual value does not

represent land value; it represents the commercial value to the developer of the site.

With such a small sample and the pilot nature of this project, the headline results

were inevitably inconclusive, inasmuch as they neither proved that innovative design

and layout improves developer value nor did they disprove it. A number of themes

emerge which, while not necessarily statistically significant, are of importance,

especially as a basis for further research validation.

Of the four locations studied, three of the four ‘exemplar’ (innovative design)

schemes yielded higher residual values than the ‘conventional’ schemes (standard

design) and one yielded poorer residual values to the developer.

The widespread market perception that good design costs more was borne out by

Executive summary


the study, but not completely. All of the exemplar schemes cost more to build per

square foot of gross internal area than the standard comparators.

The notion that higher density decreases value per square footwas not supported by this study. In two of the four cases, thevalue of the units in the higher density scheme was greater on aper square foot basis than the comparable lower-density schemes(in one case, substantially so).

Only in one scheme (Surbiton) was there a clear and definitively enhanced added

value in the "exemplar" scheme where residual value was enhanced by more than

15% over and above the standard scheme. In the case of two other exemplar

schemes (Bishop’s Stortford and Chelmsford), values appeared to be enhanced by

the nature of their design (by 7.5% and 10.3% respectively).

There appears to be no clear-cut relationship between how big a site is and whether

innovative, PPG3-led design is more profitable or not based on this small sample.

However, the clearest example of added development value was found on the

largest scheme. This may be because a large scheme has the greatest opportunity

to create its own sense of place, while smaller schemes are more vulnerable to the

influence of externalities and even other developments on adjacent sites.

Sites with relatively high residual values, in relation to their matched pairs, appeared to

have high proportions of public open space in relation to the overall size of the site.

In addition to establishing the methodology, one of the most important aspects of

the research is the identification of some possible trends in managing higher density,

better designed schemes to maximise net development value. There is a critical


9T H E V A L U E O F H O U S I N G D E S I G N A N D L A Y O U T

Residual value per area of land

Scheme Location Residual value per: Difference to matched Hectare Acre pair in same location

A Surbiton – exemplar £12.4m £5.0m +15.3%

B Surbiton – conventional £10.7m £4.3m –13.3%

C Bishop’s Stortford – exemplar £4.6m £1.9m +7.5%

D Bishop’s Stortford – conventional £4.3m £1.7m –6.9%

E Chelmsford – exemplar £3.8m £1.6m +10.3%

F Chelmsford – conventional £3.5m £1.4m –9.3%

G Aylesbury – exemplar £3.5m £1.4m –4.7%

H Aylesbury – conventional £3.7m £1.5m +5.0%

Source: FPDSavills


10

balance to be struck by the developer in maintaining unit values while increasing

density. In the schemes studied, some developers have achieved this while others

have not. Striking this balance is possibly the single most important issue in

determining the commercial value of PPG3-compliant schemes for the developer.

This alone is reason enough to advocate design practices that achieve increased

value environments, while at the same time increasing density. In these situations,

the increased value achieved is far in excess of the extra build cost incurred.

The study also raised the question of the extent to which the design and layout of a

large development scheme can affect local externalities, that is how it can actually

change the nature of a place and how it is perceived. One developer’s innovative

designs on one part of a larger development area could be a positive influence on

the value of more conventionally designed housing on another part of a site.

This research begins to illuminate the question: ‘what makes a desirable location?’

Estate agents emphasise ‘location’ as the most important element of housing value.

Leaving aside macroeconomics, there is a series of factors, including the built form,

which creates a sense of place. Some of the developers in this study have helped

to show that it is possible to create a sense of place, and therefore a new type of

location on a piece of land, by employing certain design principles. These new

places may be more desirable, and therefore more valuable, than their neighbours.

They may also be more desirable and valuable than they would otherwise have

been had standard development house types and layouts been employed. Further

research using the methodology developed in this study is urged to confirm this.

When it comes to value for a developer, as opposed to an owner-occupier, it may be more appropriate to talk about ‘density’, ratherthan ‘location’. The issues involved are complex but the rules of thegame now seem to be to maximise unit values while increasing thenumber of units on the site to meet planners’ requirements. This isnot a new game, but changing planning rules have increased thestakes. Whereas before a developer might have aimed for densitiesof between 25 and 30 per hectare (10 and 12 per acre), newplanning and design technologies have raised the target to 30 to 40per hectare (12 to 16 per acre) in suburban locations. In urbansituations, these densities would be very much higher again.

There is a strong argument that values will only be maintained at these densities if

design techniques become very much more sophisticated to maintain their

attractiveness to occupiers.

This study begins to show that good design becomes more and more critical as

densities are increased over and above PPG3’s benchmark of 30 units per hectare.

Standard design and standard layouts will not work on high-density developments.

Increases in density must accompany increased attention to design quality: there

are clear commercial rewards when this course is followed.


I N T R O D U C T I O N

1.0

Introduction


12

The Commission for Architecture and the Built Environment (CABE), with support from the

Office of the Deputy Prime Minister (ODPM) and Design for Homes, commissioned a

research study to:

1 create a methodology which could determine the relationship between housing

value and perceived housing quality and

2 show how this impacts on the developer

FPDSavills Research addressed these questions primarily in relation to betterment to the

developer. The report shows the impact of the value of different types of design to

developers by examining the influence that design has on the residual value, left after all site

development and marketing costs have been taken into account. This residual sum

potentially impacts on developer margins. It provides an approach to answering the

question of whether different types of design in housing and housing layout affect the

commercial value of new developments for a developer.

The research programme has been both experimental and practical. It provides a

methodology that can be used on any site in any location or situation to compare sites of

different types in other locations. Developers, funders, planners and other interested parties

will be able to use this approach on any site to make a like-for-like comparison between

different developments.

This report is designed to ask the question and give an indication of whether what is

deemed to be good housing design adds value to property or not. As such, it is intended to

act as a framework for further, ongoing investigation.

1.1 ContextThe issue of good design in housing has come to the fore recently in a variety of

ways. The Urban Task Force report Towards an Urban Renaissance, in 1999,

advocated the improved design quality of the urban fabric in regeneration schemes

large and small. The document concentrated on design principles in the context of

planning and policy rather than in the context of the commercial world and

development businesses.

In 2000, the DETR followed hot on the heels of the Urban Task Force report with

the, now familiar, Planning Policy Guidance Note 3: Housing, commonly referred to

as PPG3. This document broke new ground in requiring higher density new

development and setting out some of the design principles to ensure its higher

quality. Underlying this move is a shift in the planning paradigm that new

developments should be about building places rather than just predicting need and

providing the basic accommodation to fulfil that need. It was significant in

introducing detailed design principles into the planning system.

Other initiatives and movements have brought the issue of urban design, housing

design and layout to the forefront of development, architecture and planning.

Arguably, in other countries design issues have been more widely accepted and

1.0


Introduction


were taken on board earlier by mainstream, commercial developers. In the USA, for

example, the New Urbanist movement has rapidly gained acceptance and

credibility. The design, community, sustainability and urban form principles have

been adopted and endorsed in the commercial real estate world on that side of the

Atlantic for some years now. August bodies such as the Urban Land Institute

publicise and promulgate such principles. In the UK, it could be argued that these

principles have been taken up on a smaller scale and have not yet found such

widespread acceptance in the commercial world of housing development. While the

Urban Villages Forum, CABE and other bodies have subscribed to some, but not

all, of the New Urbanist design principles, these principles have been taken up by

relatively few developers. Many of the developers who have employed new design

principles have tended to be in the social housing or non-commercial sector, rather

than mainstream housebuilders.

New Urbanism is not synonymous with the best practice, innovative design now

advocated by CABE and other interested British bodies. Specifically, planned

communities do not really fit the UK context, nor does the insistence on traditional

architecture. The reference here to New Urbanism is made to highlight the contrast

with the US experience, where commercial organisations have adopted new and

different design methods on a wider scale than has hitherto been the case in the UK.

However, the situation is changing. During the period of this research project, the

authors have noted a shift in the number of development schemes where

developers have adopted at least some of the design principles of higher density,

mixed tenure, contemporary architecture and layout. An increasing number of

schemes appear to have adopted higher densities, mixed tenure and permeable,

walkable road layouts and connections.

Much of this shift is largely attributable to the increasing adoption of PPG3 at the

local planning level, forcing developers to build designs and layouts that, in most

cases, they would not otherwise have considered. One area where this new

approach is particularly noticeable is in Essex where the authors found widespread

conformity to the progressive Essex Design Guide in many new developments.

Another county where the authors noted a widespread adoption of new design

principles was in Dorset, home of the now famous New Urbanist village scheme of

Poundbury.

The increase in PPG3-compliant schemes around the country means that it is

becoming difficult in certain areas to find the more familiar, homogenous housing

estates that most people associate with modern housebuilding. It was noticeable

that in areas where ‘best practice schemes’ are to be found, planning requirements

have created a spate of similar examples and a corresponding reduction in the

development of lower density, ‘feeder road and cul-de-sac’ type layouts. However,

this was found to be a localised phenomenon. Across the country as a whole, the

authors could still identify a plethora of housing development schemes built as

single tenure, similar sized, standard housing types, arranged in cul-de-sacs, off

feeder roads adjacent to bypasses on the edge of town, built at densities of less

than 25 units per hectare.



14

It appears likely that the trend towards the more sophisticated design principles of

innovative high density has been led by some, but by no means all, local planning

department requirements rather than developers themselves, at least initially. Having

said this, it would appear that some developers have seen a commercial advantage

in developing sites according to at least some of the PPG3 principles. Discussions

with developers revealed that some increasingly believe that they can enhance

property values and, more importantly, their margins by designing and building in

these new ways. Other developers remain to be convinced.

It is exactly the proposition that certain types of design can add value that we set

out to test in this study. In order to do this, we have devised a method by which the

value of a development scheme to the developer can be measured. We are not

interested in how the capital values of individual properties may or may not be

higher than a comparably sized, standard new home. Neither are we concerned

with the amount of capital appreciation that individual homes may enjoy compared

to more standard comparisons. The concern of this study is to demonstrate what, if

any, enhanced value there is for the developer in building schemes following the

urban design principles of PPG3 compared to building more standard examples of

housing layout and design.


A D D R E S S I N G T H E G A P S

2.0

Addressing the gaps


16

2.1 IntroductionA review of the existing literature highlights a number of significant gaps, not least a

lack of empirical evidence examining the commercial value to the developer of

investing in high quality housing design in urban locations. Throughout the 1990s, a

developing body of literature highlighted a growing awareness of the issue of the

quality of urban design in the built environment, and residential design began to

attract academic attention. The publication of the report of the Urban Task Force,

Towards an Urban Renaissance (1999), gave a major policy impetus to issues of

urban design, including the design of residential development. The Urban Task

Force argued that, ‘the challenge of reducing construction costs and increasing

quality in housing is central to achieving an urban renaissance’. However, beyond

the issue of promoting good design to encourage higher residential densities and

more sustainable forms of development, the report did not examine the issue of

good design as a component of enhanced development value.

In a review of literature and case studies of commercial workspace by Bartlett in

2001, a key finding was that,

‘the research suggested that good urban design

adds value by increasing the economic viability

of development and by delivering social and

environmental benefits’.

It also stated that the existence of these benefits was increasingly acknowledged

across stakeholder groups. It was noted that good urban design can confer two

distinct forms of benefit: direct economic benefits to those investing in development,

and indirect social and environmental benefits to others and to society at large. The

researchers note that the relationship between design and value is unlikely to be a

straightforward correlation between better design and increased value, since the issue

of cost has to be taken into account. Noting the different perspectives of the public

and private sectors towards development, the research suggests that for good urban

design to be valued by all stakeholders, it needs to offer distinct dividends in the short

as well as the long term. The additional ‘value’ added by good urban design beyond

that relating to the development scheme itself is now widely acknowledged.

In arriving at any assessment of the costs and benefits of improved urban design, it

is evident, therefore, that the institutional context in which development takes place

is critical for an understanding of the relative perspectives of the different actors and

agencies involved in particular development sectors.

2.2 Housebuilding and design: institutional contextThe institutional context plays a significant role in identifying the conditions under

which the most profitable approach for housebuilders is to adopt standard designs.

Addressing the gaps2.0


The value of urban design, 2001

However, the external effects created by good design are not sufficiently well

researched, evaluated or understood by key players.

A pioneering attempt to undertake an analysis of private sector attitudes to design

and the value of it was made by Ball in 1983 in a study entitled Housing policy and

economic power: the political economy of owner-occupation. It indicated that, in

relation to the private housebuilding industry,

‘design is an important component of the

development process because, if used

appropriately, it can enhance the revenue from a

site considerably’.

Ball argues that marketability and cost minimisation tend to pull in different

directions in relation to housing design for the speculative housebuilder — although

individuality in design and non-estate site layouts may improve marketability, it tends

to raise construction costs and may lower site densities with corresponding

increases in land costs. Housebuilders minimise this potential conflict through

particular design solutions: a limited use of standard house types, with individuality

achieved by cosmetic variation of the external elevations.

Ball also notes that the market preference for detached dwellings results in a design

form which imposes cost considerations in excess of other design forms (by some

10–15%), but that this is more than offset by higher selling prices. He also indicates

that there is no acceptable definition of housing quality in the private sector, ‘so

while design has become increasingly linked with the growing market orientation of

speculative builders, associated with this trend has been a continuing fall in new

housing standards’. Ball argues that this is not primarily due to the pressure of rising

land costs, but derives from the precise institutional structure in which private

housing production in the UK takes place.

Subsequent research has confirmed that the use of standard house types has

become prevalent among large and medium-sized housebuilding companies, and

that this is driven by the institutional context and the pressures of market

competition associated with private sector housebuilding. Moreover, it has been

concluded by Hooper and Nicol that,

‘what is most striking about the overall residential

design process in speculative housebuilding is the

divorce, in design terms, between house design

and the design of housing layouts’.

Ball argues that the layout of an estate and the mix of houses implied by it are a key

aspect of private housing development, with design a central component of



Housing policy and economic power, 1983

Environment and Planning B, 2000


18

marketability rather than an aesthetic adjunct to it. Others have confirmed this

finding.

Ball’s research highlighted the fact that the competition between landowners and

housebuilders for the development gain following the grant of planning permission

had focused attention on the operation of the planning system in the post-war

period. A more recent attempt by Verhage and Needham in 1997 to specify the

particular institutional configuration which applies in the case of speculative

housebuilding in the UK, also highlighted the role of competition. It looked at the

competition between the various actors in the development process to capture

development gain and also looked at the mediating role of the land use planning

system in this process.

Verhage and Needham state that the design outcome of a development will be

dependent on a negotiation process, in which institutional arrangements regulate

the supply side of the market (i.e. housing land supply). In a similar fashion to Ball,

the authors note that, in this negotiating process, the possible gains on housing

construction do not play as important a role as the possible gains on land

development.

They argue that the land development process itself is a major influence on the

physical characteristics of residential areas. In cases where a private developer is

responsible for land development, it is a private agency that receives the

development gains, possibly sharing them with the landowner. If the costs of

secondary services are charged to the private developer by means of planning gain

agreements or other developer contributions, the remaining development gains are

reduced. In the case of private developers securing development gain, they say that

the main objective will be to maximise profit subject to realising continuity. The

authors observe, ‘the developer might try to achieve this through the quality of the

residential environment. If the products — the houses and their residential

environment — are of good quality, the developer can set its prices higher’. The limit

to this increase will be the point at which the extra expenditure equals the increase

in income resulting from the higher prices.

Verhage and Needham argue (as does Ball) that the British planning system

imposes an ‘extra price’ on housing, leading to high development gains for private

developers. In the absence of a betterment tax or levy, development obligations

have been structured to oblige private developers to spend part of the development

gain on the residential environment, and developers view these not as fixed, but

negotiable, costs.

Verhage and Needham follow through with conclusions on the implication of their

analysis. They say that, if development obligations impose additional costs, this

could lead to a decrease in production. Local authorities are therefore faced with a

dilemma in that promoting a larger supply of housing could result in a lower quality

of residential environment. Similarly, promoting a higher quality residential

environment could result in a reduced supply of housing.

Leopold and Bishop, in their 1983 report Design philosophy and practice in


speculative housebuilding, examined the consequences of this institutional context

for the process of housing design. They argued that the speculative element in land

development gain in private sector housebuilding requires a type of housing which

minimises the special high risks associated with this form of housing provision.

The existence of a portfolio of such designs enables developers to make a

reasonable estimate of costs of production at short notice, and to make a rapid

estimate of the potential ‘yield’ of any site. The authors observe that, ‘not

surprisingly, housing design, in this context, plays an inherently conservative role,

one that inhibits any radical approach to housebuilding that might operate through

the design process’. Booth, among others, has noted this conservatism in design to

be a long-term feature of the housebuilding industry. Moreover, Leopold and Bishop

assert that, ‘the combination of standardised, fairly stable individual house plans

with very flexible, market-sensitive elevations and trim is the physical expression of

the compromise between the long-term stability of the production process and the

more volatile demands of the market’.

The most recent review of housing design quality, Housing design quality: through

policy, guidance and review by Carmona, supports these general conclusions. It

argues that, ‘without some public sector intervention, the unique market

circumstances in which housebuilding occurs will continue to ensure the widely

accepted (even among many housebuilders) devaluing of design in the residential

sector’. Despite this, some commentators have noted that planning intervention in

the form of design guidance may have a positive impact, creating a product with

enhanced sales values and achieving development at higher densities.

In some instances, the increased certainty afforded by coherent residential design

guidance is argued to be of potential assistance to housebuilders. However, it is

also noted by Carmona that pressure from landowners for a share of development

gain results in a situation in which

‘expectations are conceived in a design vacuum, with

— usually — no consideration given to the cost of

delivering higher quality development’.

As a consequence of this pressure, housebuilders often have to strategically ignore

local authority design aspirations in order to make their bid. In order to avoid this,

some commentators have argued that the only means of increasing residential

design quality is to access the development value for investment in design.

2.3 Brownfield land development and housingThe literature on design and housebuilding is limited in that, either explicitly or

implicitly, it reflects a predominantly greenfield context for housing development. As

the scope for such development has become constrained by major Government

policy initiatives (most notably PPG3), and particularly the sequential approach to



Housing design quality, 2001


20

urban land release, the potential of previously developed land for housing has

received increased attention.

Research into the costs of residential development on greenfield sites appears not

to have been undertaken for brownfield development. A study by Conran Roche

and Davis Langdon & Everest in 1989 broke down residential development costs

but did not deal explicitly with design issues.

Literature on brownfield development land by authors such as Syms presents the

argument that the development of brownfield sites may be characterised by high

levels of financial uncertainty. This is particularly evident in the context of the loss of

English Partnership’s ‘gap-funding’ regime. Financial uncertainty is exacerbated in

the context of previously developed sites that are affected by land contamination

which requires remediation. Such uncertainty may significantly affect the context of

negotiation between housebuilding companies and local planning authorities, in

which improved design may be only one issue among many competing, material

planning considerations. In a review of progress relating to the identification of

brownfield housing land, research by the Civic Trust in 1999 found a poor fit

between the original identification of sites for housing and the actual achievement of

successful housing development on them. Again, there was no specific reference to

design issues in this research (although adverse market conditions and the

unwillingness of landowners to sell feature prominently in it).

Successive studies by Llewelyn-Davies have investigated the possibilities of

increasing the housing capacity of urban land. A major conclusion has been that,

‘there is potential to significantly increase residential densities and at the same time

improve the environmental quality of new residential development. This can be

achieved through a creative design-led approach which responds to the particular

characteristics of a site, its surroundings and the needs of future residents’. The

essence of this approach is that the objective of quality is addressed through design

rather than by applying general planning standards.

The authors state that an important consequence of this recommendation is that,

‘developers must invest more in design to achieve

high quality development at higher densities. This

means less reliance on standard types and a

greater freedom for designers to respond to the

potential of individual sites’.

No explicit economic analysis of the costs and benefits of such an approach was

undertaken.

Others, such as Rudlin and Falk, offer a different perspective. They advocate both

urban renaissance and improved housing design. Writing in 1999, they say that, ‘in

both the private and social housing fields there are powerful forces at play which

reinforce both suburban designs and suburban locations as well as exerting a


Llewelyn Davies, 2000

corrosive effect on quality and resisting innovation’. They go on to indicate that,

‘while there are signs that things may be changing, these forces must currently be

seen as a major constraint on the promotion of more urban housing’.

A cautionary note is sounded in other research regarding the operation of the land

and housing market under conditions of supply constraint in the UK. Cheshire and

Sheppard in 1997 indicated that, under conditions of sustained housing demand and

land supply constraints, it is generally ‘planning amenities’ or space that increase

disproportionately in price in comparison with ‘reproducible’ attributes of housing.

These ‘goods’ which the planning system delivers are valued, the authors argue, but

in ignorance of the price that is paid for those planning amenities. ‘Their values are

not just incorporated in the price paid for a house. There are also extra costs. These

are the higher price paid for all housing land, shortage of space and the more

cramped conditions in which we all have to live. Planning amenities are not paid for

directly so they appear to be free. But because they are only produced by restricting

the supply of building land, they cause the price of land, and thus of housing, to rise.’

This analysis would seem to confirm and extend that of Verhage and Needham in

identifying the source of the ‘extra price’ attributable to the ‘residential environment’

outside the dwelling unit in the operation of a restrictive planning system. Design is

not, however, one of the explicit housing attributes considered in the analysis

(although access to schools and to public open space is modelled). Others also

make a similar argument concerning such ‘neighbourhood externality’ effects of

planning-induced improvements in the quality of residential development.

2.4 Housing design quality and ‘value’The publication in 1973 of A design guide for residential areas by Essex County

Council introduced a radically new framework for residential design guidance.

Subsequent research into the impact of such design guidance for private residential

development by the DOE/HRF in 1976 indicated a concern with the cost

implications of design guidance and control. It asserted that, ‘a requirement for a

more expensive design solution for the houses must be offset by a saving in cost

elsewhere, such as in road construction, or by allowing a higher density of

development to reduce the land costs for each dwelling’.

In a discussion paper on good design in housing by Davison in 1990, it is argued

that many aspects of good design add little or no extra costs to residential

schemes, but that some elements of good design do increase costs. It is argued

that, ‘developers who have persisted with high design standards have found that

their product does have an added, premium value, but only on the subsequent

resale when the environment is more settled and established. This benefits

discerning purchasers but does not recoup the developer’s costs’.

Davison estimates that the cost of achieving a broad spectrum of design

improvements might be in the order of 5–10% of the land value. He argues, ‘this

seems to suggest that there should be some scope for planning intervention to

redirect some of the betterment from the landowners’ profits to the benefit of the




22

local community with better design, materials and open space’. It is, nevertheless,

claimed that intervention of this type, using planning agreements, is difficult to

achieve, except on sites not already allocated in a local plan.

In a review of the role of housing standards in promoting housing quality in 1992,

Karn argued that ‘not only are there no financial inducements to provide high

standard new housing, there are no national targets for housing quality’. He added

that, ‘certainly there is now a groundswell of unease about trends in standards and

quality of design’. Subsequent research by Karn and Sheridan in 1994 indicated the

very different parameters that influence housing standards and design quality in the

private and housing association sectors. They indicated that, ‘to improve quality of

design in the private sector, we, therefore, conclude that the best mechanism is to

strengthen consumer pressure’.

The Government discussion document Quality in town and country in 1994 argued

in its preface that,

‘Quality pays. Good quality is good economics’.

However, no argument or research was cited to demonstrate this assertion. In

1998, a DETR report examining ways to improve the quality of new housing layouts

argued that many of the obstacles to enhancing quality are symptoms of land

valuation that ignores the cost of quality and the lack of cooperation in the design

process. The report argued, ‘bad design can increase land values. Over-

development with uniform house types laid out on minimum plots, in schemes that

ignore site characteristics and local identity, will maximise the land value. Good

design does not necessarily increase costs, but it is rarely able to match the land

value achieved by bad design. It is therefore essential that quality targets are

established before land values become fixed’.

The report also noted the role of standard house types in maximising the

development value, and therefore the land value, which is considered an essential

aspect of layout design in a competitive market. The limitations of these standard

house types on urban redevelopment sites are recognised and the perspective of

the developer is regarded as crucial. ‘Developers look within the site to assess the

value of the land and the commercial viability of their investment. The importance of

this aspect of development cannot be overstated; it is largely through the

investment of development companies that plans are realised. But commercial

considerations, which should run in parallel with quality and sustainability objectives,

are allowed to dominate the developer’s thinking, often to the exclusion of other

matters. The developer’s appraisal of the setting of the site is concerned with the

housing market rather than design opportunities for the development. Developers

must be encouraged to take a broader view of their activities’.

Research into the external appearance and site layout of new houses,


Quality in town and country, 1994

commissioned by the Popular Housing Forum in 1998 indicated that, for owners of

new-build housing, ‘issues of site layout were considered more important than the

appearance of an individual property, especially the area in which a house was built.

The appearance of the neighbourhood was seen as more important than the

appearance of the individual house’.

Carmona argued, in 1999, that the issue of the dominance of roads in the

determination of housing layouts is arguably the most difficult component of the

residential landscape to integrate successfully. Not surprisingly, then, the current

period of residential design guidance was ushered in with the publication by the

DETR in 1998 of the report Places, streets & movement, a companion guide to

Design Bulletin 32 — Residential roads and footpaths. In the same year, the

publication of Planning and development briefs: a guide to better practice endorsed

the need to consider design issues at the earliest stage of the planning and

development process (a point which was made strongly in the 1998

POS/HBF/DETR report).

The publication by the DETR of Planning Policy Guidance Note 3: Housing in 2000

consolidated the transformation in national planning policy guidance relating to

design which Carmona dates from the original guidance issued in 1992. The

publication of By design — urban design in the planning system: towards better

practice by DETR/CABE in 2000 began the process of providing national guidance

relating to urban design objectives and aspects of development form. English

Partnerships and The Housing Corporation published the Urban design

compendium by Llewelyn-Davies in 2000 as a complement to By design to

contribute to the improvement of the quality of housing-led regeneration projects

and the promotion of sustainable new development. A companion guide to PPG3,

By design: better places to live, published by DTLR/CABE in 2001 sought to apply

the principles of good quality urban design (set out in Places, streets and movement

and By design) to the residential environment. None of these official publications

considers in detail the issue of costs and added value from improved design quality

in residential environments.

2.5 SummaryThe most striking finding in a review of the literature relating to the quality of

residential design is the almost complete absence of any empirical attempts to

measure the implications of high quality on costs, prices or values.

Few commentators have gone beyond making a distinction between the direct

benefits for built environment stakeholders and those indirect benefits that accrue in

the form of ‘public goods’ or externality effects. Initial attempts to formulate models

of the residential development process which emphasise the institutional and

negotiating context in which residential design solutions are formulated have tended

to be based in a greenfield development context, with few applications in brownfield

contexts. To the extent that brownfield developments involve complex pricing effects

deriving from public goods and externality effects deriving, in part at least, from the




24

operation of the land use planning system, the attribution of relative costs and

benefits deriving from design per se have been difficult to isolate and measure. In

particular, it becomes difficult to isolate ‘value’ attributes of residential design

deriving from the dwelling from those deriving from its locational context.

Those who have focused on the process of residential design (at least in the private

sector) have emphasised the crucial importance of the institutional framework in

determining the parameters within which different actors and agencies carry out

housing development. In this process, the establishment and apportionment of

‘added value’ deriving from design is seen to be partially contingent on the place in

the development process at which crucial decisions on design are made (which are,

in turn, reflected in residual land values). The apportionment of that part of the

‘added value’ deriving from design in the development process is dependent on the

complex operation of housing and land markets, so that part of this ‘added value’

may appear upstream or downstream of the residential development process itself.

Very little research has addressed the issue of the commercial value of housing

design to the developer. Most literature relating to architecture, planning, housing

and economics only touches on the issue of the development value that results

from design in the most marginal fashion. Not surprisingly, therefore, methodological

discussions are virtually non-existent. Where they do occur, they usually relate to

greenfield development. In the very considerable literature dealing with urban design

in general, and residential design in particular, there is very little reference at all even

to cost implications, let alone ‘value’.

The references cited in this literature review are contained in Appendix 1.


3.0

Methodology

M E T H O D O L O G Y


26

3.1 OverviewThe aim of this study was to define how different types of housing layout and design may

affect the costs of development and the value of the end-product and, hence, how

developer profitability might be affected. The following questions were addressed:

■ What constitutes good design?

■ How can economic value be determined?

■ How can externalities be dealt with?

The research programme comprised three parts.

1 Investigate the background to the project, conduct necessary literature reviews and

justify the chosen empirical methodology involving the selection and comparison of the

matched examples

2 Select appropriate examples of completed developments in matched pairs to

compare developments with certain innovative design elements to those of standard

design. Put a cost and a market value on the developments to determine whether

there is any difference in a standard measure of residual value and therefore potential

developer margins

3 Examine the findings on the selected projects and discuss what forces and factors

may be behind the results. Compare these to other research findings, if appropriate,

and recommend any further research work that might be needed

In overview, a method was devised to compare developments on a like-for-like

basis to facilitate the comparison of different schemes. This was done by comparing

a series of eight examples of different types of housing in four suburban locations

within the South-East. Of these developments, four (one in each location) could be

described as largely PPG3-compliant, in that they employ some degree of

innovative design, and were selected by the commissioners of the report to

represent best practice examples (these have been termed ‘exemplar’ in the report).

The remainder were good developments but following a more standard design

(these have been termed ‘conventional’ in the report). Two of the ‘conventional’

developments in this study are hypothetical as no cooperative developers on a real

development could be found to act as comparators to the exemplar schemes.

The issue of determining commercial value was done on the basis of a residual (i.e.

the sum left after deducting all onsite development, construction and marketing

costs from the house sales revenue). It must be emphasised here that, despite

being expressed as a residual sum per area of land, this residual does not represent

land value. The residual covers land costs, corporate overheads and developer

margin. It also, importantly, represents the sum available to the developer to carry

out any section 106 or other agreements to build community facilities and any

affordable housing element of the development, for example. While the cost of

these agreements is often critical in determining developer margin, we have not

measured them. This is because they are negotiated on a site-by-site basis and

Methodology


3.0

therefore vary greatly from scheme to scheme. Their inclusion in cost measurement

would not therefore facilitate a like-for-like comparison to be made between

schemes. The residual sum measured in this research is intended to represent both

the commercial value to the developer of the site and also some of the social value

since, without it, the developer could not afford to provide any loss-making social

amenity buildings.

It must be noted here that, given the way the development market works, some of

the residual value measured in this report would be shared with the landowner.

Exactly how much depends not only on what proportion of the residual sum is used

to provide social housing and other section 106 buildings and works, what is

needed in the form of central company overheads, but also on the developer’s

negotiating position. This will be determined by factors such as the degree of

competition for the site from other developers and the form of contract that the

developer enters into with the landowner. No attempt was made to assess how

much of the residual would end up in the pockets of the landowner rather than the

developer.

The methodology was designed to compare each site on a like-for-like basis so all

valuations were made at a single point in time (March 2002) and as if all units were

being sold on the same (exemplar) site in each location. The measurement of land

area was also standardised.

A simple, step-by-step guide to the methodology used is contained in Appendix 4.

Further details of this methodology are given below.

3.2 Selection of case studiesThis research used case studies of real development schemes wherever possible to avoid

being unduly theoretical. The case studies in this research were selected on the basis that

they represented a range of design and layout principles. Four different locations were

selected, all of which contained examples of schemes that followed the urban design

principles in PPG3 or innovative housing developments.

Although initially a variety of locations across the country was identified for study, it was

decided that it would be difficult to identify any differences in value that were due to

regional externalities and regional market differences rather than the issue of design. It

was therefore decided that all locations should be in the South-East to eliminate any

differences attributable to variations in regional house markets and regional economic

considerations.

The sites selected were all on the edge of towns or district centres and all the sites were

suburban in nature. They were Surbiton in outer south-west London, Bishop’s Stortford in

East Hertfordshire, Chelmsford in Essex and Aylesbury in Buckinghamshire.

All of the development schemes selected were examples of mainstream, mass house

building of a good build standard. In each location, two schemes with different types of

design and layout were selected. One was selected on the basis that it was largely

PPG3-compliant and an example of good design practice. (Subsequently, on careful




28

measurement, some of these schemes turned out not to be PPG3-compliant on the

strict measure of minimum density.) The best practice sites were termed the ‘exemplar’

site in each case. The other site in each location was selected on the basis that it was a

typical example of a standard design and layout type in that location. These sites were

termed the ‘conventional’ site in each case.

The following criteria, established in By design, the DTLR/CABE companion guide

to PPG1, were used to select the exemplary schemes.

■ Character — a place with its own identity

■ Continuity and enclosure — a place where private and public spaces are

easily distinguished

■ Quality of the public realm — a place with attractive and successful outdoor areas

■ Ease of movement — a place that is easy to get to and move through

■ Legibility — a place that has a clear image and is easy to understand

■ Adaptability — a place that can change easily

■ Diversity — a place with variety and choice

Schemes that broadly conformed to these characteristics were selected as the

‘exemplar’ schemes.

The conventional schemes were selected on the basis of their proximity to the

exemplar site and that they represented a good example of what could otherwise

have feasibly been built on the exemplar site. The sites were chosen on the basis of

no other characteristics than that they represented nearby conventional residential

schemes of a decent quality, likely to appeal to broadly the same market sector (in

terms of price) as the exemplar scheme.

A major problem in finding suitable case studies was the reluctance of developers to

participate in the study. To form comparators for the four exemplar schemes,

suitable ‘conventional’ schemes were identified in each case but two of their

developers later refused to participate in the study. Two hypothetical, standard

development schemes were created and costed in the two locations that were left

without ‘conventional’ sites.

There were several reasons for developers’ unwillingness to participate, the main

one being confidentiality. There was also a fear of being held up as an example of

bad design. The ‘conventional’ sites are put forward as reasonable examples of

conventional development with standardised layout practices. Their only shared

characteristic is that, broadly, they are either not PPG3-compliant to the same

extent as the exemplars or that they do not exhibit the innovative or best practice

design principles, set out in the methodology above, to the same extent as the

exemplar examples.

It is also worth noting that in the areas studied there was an increasing scarcity of

non-PPG3 compliant, standard sites to act as conventional examples.

Development, in the South-East and other southern parts of the country at least, is

increasingly characterised by PPG3-type design principles.


3.3 Dealing with externalitiesOne reason put forward for the lack of comparable studies of value between good design

and standard design quality is the fact that it is difficult to separate the differences in value

that are due to design and those that are due to other factors. These other factors may

include, among other things, access to schools, access to transport, access to other

amenities, etc. Some research studies have attempted to account for these externalities

through sophisticated statistical techniques such as hedonic mix adjusting methods and

the validity of this methodology is addressed in the literature review section of this study.

For the purposes of this study, we originally proposed to identify matched pairs of

‘exemplar’ sites and ‘conventional’ sites with as near identical external characteristics as

possible. The impact of externalities would then be removed in two ways:

1 through eliminating as many of the differences between sites’ locations as possible

2 by measuring the impact of development on a homogenous product, namely land,

rather than trying to assess the difference in value between exogenous units

An initial attempt was made to eliminate site differences through the utilisation of

geographical information systems. However, it was found that there are simply not

enough new build schemes of sufficient size in comparable locations to yield comparators

on this basis.

A revised alternative method of excluding the impact of external influences on value was

required. Instead, conventional developments in the same area but sometimes in slightly

different types of location were chosen. By using a valuation approach to assess the

gross development value of all schemes, the influence of different external factors on each

scheme was then eliminated. This was done by asking experienced valuers, active in the

local market-place, to assess each scheme as if it were on the same site as each of the

exemplar schemes. In this way, like-for-like comparisons were obtained. The only variable

that differed between each matched pair was the design and layout.

During the course of the research, two of the developers of conventional, comparator

schemes decided to withdraw from the study. These schemes (at Chelmsford and

Aylesbury) were replaced by the creation of a hypothetical scheme in each area. These

hypothetical sites contained similar characteristics to the sites already selected (to provide

a grounding in reality). A cost model of these sites was then created by Davis Langdon &

Everest (DL&E) to emulate a typical, conventional local scheme.

The focus of this study on residual value per area of land, as opposed to unit value or

value per square foot of unit, provided a common basis of measurement for all the sites. It

was also a method by which they could all validly be compared. This negated the need to

make adjustments or account for variations in the charateristics of the units on which

value was measured. Instead, the residual value could be specified on a homogenous,

uniform product (land) which differed only by the location of each matched pair. Even this

broader, regional difference between the matched pairs could be eliminated by indexation

using an appropriate index of regional building land values, if required.




30

3.4 Measuring and quantifying the sites3.4.1 Site area

It became apparent during the course of this study that different developers

measure site area in different ways. It was therefore necessary to standardise the

measurement of sites in order to determine costs and values by land area on a like-

for-like basis.

It was decided that the standard measure should be the total land area that would

have to be bought by the developer in order to complete the development but

excluding:

■ any areas of major infrastructure such as bypasses and any associated

landscaping, major distributor roads, etc

■ land for section 106 development such as social housing, schools, pavilions, etc

■ public open space including verges, footpaths not adjacent to roads (not

pavements), cycle paths, village greens, recreation grounds, parks and other

open space

Each scheme was measured using online Ordnance Survey (OS) maps or by

digitising detailed OS site plans where the OS map was not available online. Where

site boundaries were divided by roads, measurement was made up to the centre of

the carriageway unless there was no current or future development on the other

side of the road, in which case the whole carriageway was included.

While these measures may not always accord with what was done in practice on a

site, it ensured that every scheme has been measured in the same way.

3.4.2 Layout type

The layout characteristics of each scheme were observed and described. Empirical

measurement of the proportion of site land area utilised for different uses was also

made. These uses included:

■ public open space

■ peripheral and feeder roads

■ access roads and cul-de-sacs

■ public driveways, parking courtyards and community space

■ private driveways

■ garages

■ private gardens

■ houses and flats

The proportion of land used for each of the above was intended to give a clearly

quantified picture of the different nature of each scheme.


3.4.3 Appearance

Visits were made to each site and records made of the appearance of buildings and

the site in general. Marketing material was also gathered to give a picture of internal

layouts, appearance and finishes. In the case of the hypothetical schemes, detailed

specifications, appearance and layout were devised.

3.5 Measuring costsThe building and marketing costs of each scheme were measured and assessed by

Davis Langdon & Everest (DL&E) in extensive site visits and interviews with developers.

In the case of the hypothetical sites, detailed development cost models were utilised in

order to determine the cost to the developer of building and marketing a site.

The methodologies employed are detailed in Appendix 2, which contains the DL&E

report.

The principle used in all the costings was that they should include all external works

required to develop the plots, professional fees, sales and marketing expenses. The

costs were deliberately compiled to exclude the costs of strategic site infrastructure

such as remedial works and major roads, bypasses, road screening, etc. It also

excluded the costs of any section 106 and 278 works as well as funding costs and

central, company overhead contributions.

Costs were stated at March 2002 price levels, having been adjusted using cost

indicators, where appropriate. The costs reflect the locality of each pair of projects

and were not adjusted on a regional or local basis. The cost levels all assume the

completion of the scheme as a single phase for sale in March 2003.

3.6 Measuring valueLittle research has been conducted to ascertain the relative economic value of good

design for the developer. This was specifically addressed by looking at the value of

the whole scheme, rather than individual units (and by looking at the total

development costs that might be involved) in order to determine economic value to

the developer. The reasoning behind this approach was that it eliminated some of

the effects of externalities and looked at commercial value rather than value to the

end-user, the owner-occupier.

Approaches that have been used in other studies (for example, Eppli and Tu,

Valuing the new urbanism, Urban Land Institute, 2001) have looked at value from

the point of view of the open-market value:

■ of individual units

■ traded by, and to, owner-occupiers

■ over time

The disadvantage of this approach was that it does not illustrate the value (or




32

otherwise) of design to the developer (either an organisation or individual) who will

actually be responsible for building the units (to any given design standard). It is

developers, not owner-occupiers, who will determine whether certain design

practices are put in place or not.

The examination of individual unit values also created problems of comparison

between units, which varied greatly when ownership patterns and other factors

differed. Attempts to adjust for these factors through hedonic mix adjusting were

seen as inadequate because they could never account for every possible influence

on value. Similarly, the choice of factors for which adjustment was made was, at

best, subjective and often arbitrary. Details of the calculations made in this study are

contained in Appendix 5.

A further problem was that other studies have used real transaction data to

determine value differences. This was dangerous because the type and nature of

properties in these transaction samples alter over time so it has not been possible

to compare like with like at every stage. It was considered that there was no

advantage in knowing that house type A sold for x and house type B sold for x + y

if the two could not be validly compared and/or they were sold at different times.

Likewise the examination of price growth failed to account for externalities that may

have been due to factors such as differential economic performance between

localities, the presence of nearby blight, changes in infrastructure and other external

causes which may have been difficult either to account for or eliminate.

Another possible method of approaching this research would have been to use a

contingent valuation method. This would have involved appraising an ‘exemplar’ site

on which best practice housing had been built but then assuming that it had been

built to a ‘standard’ specification, layout and design. The resulting hypothetical value

could then have been compared with the actual values achieved on the best

practice site and conclusions could have been drawn. This approach would have

been valid to an extent (and indeed would have been very similar to the approach

that was actually used where the conventional sites were hypothetical) but it was felt

that such a completely hypothetical approach would not have been entirely credible

to all observers.

The total development value of all the residential units in the schemes under study

was assessed. In order to do this, a variety of local estate agents, valuers and

development sales experts were contacted and asked to value each unit in every

scheme. This method was imperative in the case of the hypothetical examples in

any case but also had the advantage of eliminating any need to adjust revenues

over time as all valuations were made for March 2002.

This valuation method relied on the skill and judgement of valuers who were making

a subjective and hypothetical assessment. Care was taken to ensure that the

valuers used were highly experienced in the field and in the local area with

experience of either selling on the scheme or valuing similar schemes. In addition, at

least two, and often three, opinions of value were sought from different valuers in


each case. The results were checked for anomalies and the average of all valuations

for each unit was used.

In the case of the hypothetical examples, the schemes were carefully put together

to reflect real, newly built properties that actually exist in the location in question.

Consequently, the valuers were valuing unit types for which they had real

comparables and real-life experience.

3.7 Assessing commercial valueThe assessment of the commercial value to a developer of the scheme was made by:

1 taking the total gross value of the residential units only, as described above

2 subtracting the costs of developing the plots on which the units stand, building

the units, finishing and marketing them according to the criteria set out above

3 analysing this on a per hectare of standardised land measurement

It is important to note that the resulting residual sums in each case never would

have been representative of land value. They would be representative of sums

available to the developer to cover both land costs and central corporate overheads

as well as to be available as developer margin and profit (or loss) on the site.



4.0

Research results and discussion

R E S E A R C H R E S U L T S A N D D I S C U S S I O N


36

4.1 Commercial valueWith such a small sample, the headline results were inevitably inconclusive

inasmuch as they neither prove nor disprove that innovative design and layout

improve developer value. The methodology piloted in this report is itself a significant

contribution to understanding values.

Of the four locations studied, three exemplar schemes yielded higher residual values

than the conventional schemes and one yielded poorer residual values to the

developer. In this one case, the difference in the residual value per hectare of land

was less than 5%. Only in one scheme (Surbiton exemplar) was there a clear and

definitively enhanced added value in the exemplar scheme. Here, residual value was

enhanced by more than 15% over and above the standard scheme. In the

exemplars at Bishop’s Stortford and Chelmsford, both schemes’ values appeared to

be enhanced by the nature of their design (by 7.5% and 10.3% respectively).

The following commentary provides some analysis as to what might be behind

these differences in residual values.

4.2 Scheme characteristics The tables in Appendix 3 contain the detailed characteristics of all the sites which help to

explain the different schemes’ performances. The differences between all the schemes

are many, sometimes subtle and certainly varied. Not all of the exemplar examples are by

any means similar, and even the standard examples vary from each other substantially.

4.2.1 Site size

Determining the size of each site in the study proved to be one of the most

challenging aspects of this work. It would appear that the method by which

Research results anddiscussion


4.0

Scheme Location Residual value per: Difference to matched Hectare Acre pair in same location

A Surbiton – exemplar £12.4m £5.0m +15.3%

B Surbiton – conventional £10.7m £4.3m –13.3%

C Bishop’s Stortford – exemplar £4.6m £1.9m +7.5%

D Bishop’s Stortford – conventional £4.3m £1.7m –6.9%

E Chelmsford – exemplar £3.8m £1.6m +10.3%

F Chelmsford – conventional £3.5m £1.4m –9.3%

G Aylesbury – exemplar £3.5m £1.4m –4.7%

H Aylesbury – conventional £3.7m £1.5m +5.0%

Source: FPDSavills

Table 4.1: Residual value per

area of land

developers measure their sites is by no means standard and varies quite

substantially. Some appear to be very conservative in their measurements (perhaps

to raise the apparent density on the site) while others include every piece of land

bought in order to facilitate the development.

It became apparent that the method by which each and every site was to be

measured needed to be standardised. The standardised measurement is similar to

that outlined in PPG3 but without distinguishing between children’s playgrounds and

other public open space. When measured this way, the schemes studied vary in size

from under 1.5 hectares (under 4 acres) to over 7 hectares (over 18 acres).

The sample contains a wide variety of site sizes ranging from small stand-alone

schemes through small phases on large schemes, big phases on larger schemes

through to large standalone sites.

There appears to be no clear-cut relationship between how big a site is and whether

innovative design is profitable or not. The sample size is insufficient to determine

whether the size of the site itself is an important factor in determining the commercial

performance of different types of design on the site studied. It is notable, however,

that the clearest cut example of added development value was found on the biggest

scheme. This may be because a large scheme has the biggest opportunity to really

create its own sense of place while smaller schemes are more prone to the influence

of other externalities and even other developments on adjacent sites.

It could be that the crucial factor is not size, per se, but the extent to which a

scheme is self-contained. It is notable, for example, that the conventional scheme at

Bishop’s Stortford is small but it is self-contained and distinctive. Due to the

characteristics of the location at Chelmsford, on the other hand, there are fewer

differences between the hypothetical, conventional scheme and the exemplar

scheme. They are not separated or distinct but both conform to very similar layouts

and both are Essex Design Guide compliant.

It is possible that the image of place that has been created by the exemplar schemes

in each case has influenced the value of adjacent conventional schemes. This



Scheme Location Total area Standardised area Developer’s figure

Hectares Acres Hectares Acres Hectares Acres

A Surbiton – exemplar 9.1 22.6 7.3 18.0 8.3 20.5

B Surbiton – conventional 4.4 10.9 3.9 9.7 4.9 12.2

C Bishop’s Stortford – exemplar 2.7 6.7 2.6 6.4 3.1 7.5

D Bishop’s Stortford – conventional 1.5 3.8 1.5 3.7 1.5 3.7

E Chelmsford – exemplar 4.1 10.1 3.5 8.6 3.5 8.7

F Chelmsford – conventional 4.2 10.4 3.8 9.4 N/A N/A

G Aylesbury – exemplar 1.8 4.6 1.8 4.4 1.9 4.6

H Aylesbury – conventional 2.9 7.3 2.3 5.8 N/A N/A

Source: FPDSavills

Table 4.2: Site size, by

location


38

possibility reveals a possible flaw in the methodology. Because all the conventional

schemes were valued as if they were in the same location (although not on precisely

the same site) as the exemplar schemes, an important component of design value

may not be illustrated. This research will not pick up the possible added value to a

place that innovative design might bring.

It is possible that some of the conventional schemes might benefit from being

located adjacent to a successful exemplar scheme and achieve higher values than

they might otherwise achieve in other surroundings. There may be future

methodology revisions and designs that could overcome or even test this factor,

perhaps to measure how an exemplar scheme might improve locally held

perceptions of a new address, notably where it forms the frontage to the perimeter

road for more standardised estate-type development behind.

4.2.2 Site density

Density is central to discussions on design and layout and is key to PPG3 but it is often

poorly understood and crudely applied. It became apparent, within the parameters of

this research, that density could be substantially affected by measurement norms and,

even when measurement is standardised, can yield some surprising results.

It was clear that different schemes were built at different densities, for a variety of reasons.

The conventional planning measure of density, used in PPG3, is in terms of units per hectare.

By this measure, and using the standard site measurement that excludes public open

space but includes access and minor perimeter roads, there was a range of site densities

— from 16.2 units per hectare (6.5 per acre) to 43.3 units per hectare (17.5 per acre).

Table 4.3 shows the unit density per hectare and per acre for each of the sites

studied. It shows a surprisingly wide range of densities between sites and, most

surprisingly perhaps, between the exemplar sites. In fact, two of the exemplar sites

(Bishop’s Stortford and Chelmsford) turned out not to be PPG3-compliant in this

respect because they had less than 30 units per hectare.


Scheme Location No. of units per:Hectare Acre

A Surbiton – exemplar 43.3 17.5

B Surbiton – conventional 29.5 11.9

C Bishop’s Stortford – exemplar 22.3 9.0

D Bishop’s Stortford – conventional 16.2 6.5

E Chelmsford – exemplar 27.6 11.2

F Chelmsford – conventional 26.5 10.7

G Aylesbury – exemplar 41.3 16.7

H Aylesbury – conventional 32.5 13.2

Source: FPDSavills

Table 4.3: Density: units per

hectare (acre)

It is interesting to note that, although all the conventional schemes are of lower

density than their matched pairs, some are of higher density than other exemplar

schemes elsewhere. It is noticeable that the density of units per hectare is over twice

as high for the Surbiton exemplar (scheme A) than for its matched pair. This unit

density was achieved by building apartments on the site and it reflects the fact that

this location is among the most urban of all the examples. Achieving similar densities

by building apartments on the scale of scheme A would simply not be an appropriate

option in most of the other locations. It seems certain that it is this very high density

that has contributed to the high residual values seen at the scheme A site.

In this respect, PPG3 can be the developer’s ally by allowing a higher number of

units to be built on a site than might otherwise have been the case in the past. This

will enable unit production to rise on individual sites. Whether this would lead to

higher overall production of units is subject to other factors such as land availability

and is a potential subject for further research.

There is also a balance to be sought between density and the perceived quality of a

place. The Surbiton exemplar (scheme A) arguably achieved this balance by

combining dense and efficient terraced housing and apartments with large areas of

open space. This made for an attractive location coupled with very efficient land use.

It is worth noting the urban nature of such a solution compared to what might be

possible on edge of town sites. Also, the size of the scheme, with its availability of

large open spaces, may be critical in successfully achieving these densities without

overly damaging market value. In this instance, the units in scheme A were worth

slightly less per square metre than its matched conventional scheme, B. It was the

higher number of square metres of built space accommodated in each hectare that

made the big difference in residual land value.

4.2.3 Development density

Development density differs from site density in that it describes something about the

volume of houses built as well as the numbers built for each area of land.

Development density is expressed as a ratio. The ratio is a product of the number of



Scheme Location Total area of Coverage of housing

development: ft2 units: ft2

A Surbiton – exemplar 785,393 442,786

B Surbiton – conventional 424,172 188,266

C Bishop’s Stortford – exemplar 280,747 93,947

D Bishop’s Stortford – conventional 159,991 68,243

E Chelmsford – exemplar 374,038 121,166

F Chelmsford – conventional 409,916 121,916

G Aylesbury – exemplar 192,991 59,690

H Aylesbury – conventional 251,538 73,364

Source: FPDSavills

Table 4.4: Site coverage, by

location


40

square metres (or square feet) of gross internal floor area divided by the number of

square metres (or square feet) of site area.

While this figure alone is unable to describe how ‘clustered’ the development is, in

most cases, the exemplar sites achieved higher build ratios than the conventional

schemes. The exception was the Bishop’s Stortford conventional scheme, D, which

achieved high development density ratios by putting rooms in the loft and thereby,

effectively, building to three storeys.

Scheme A illustrates that raising building height is the most effective way to achieve

higher development density. Again, there is an issue with how appropriate taller forms

might be in all rural, suburban and urban situations.

4.2.4 Land use

One of the clearest differences that could be measured between schemes was the

different proportion of land use in each. An analysis of land use differentials was

possibly the single most important tool to distinguish the different nature of design

and layout between schemes. The tables in Appendix 3 show the proportion of land

taken up by different land uses.

It can be seen that sites with relatively high residual values in relation to their matched

pairs also appeared to have high proportions of public open space in relation to the

overall size of the site.

4.2.5 Unit size

In nearly all cases, exemplar schemes had smaller unit sizes than the conventional

schemes. This may be because higher densities have placed more emphasis on the

size of the public realm rather than the size of the private realm.


Scheme Location Development density ratio

A Surbiton – exemplar 0.564

B Surbiton – conventional 0.444

C Bishop’s Stortford – exemplar 0.335

D Bishop’s Stortford – conventional 0.427

E Chelmsford – exemplar 0.324

F Chelmsford – conventional 0.298

G Aylesbury – exemplar 0.310

H Aylesbury – conventional 0.292

Source: FPDSavills

Table 4.5: Density: gross

internal area (gia) as

percentage of site area

Generally speaking, a bigger unit will sell for more than its smaller counterpart in the other

matched pair. However, the residual value calculations in this report show that this increased

value does not always make up for the more land-hungry nature of a larger building.

4.3 CostsBuild costs are generally considered by housebuilders to be a very important

component of the house building process. There are few developers whose interest

is not closely focused on this element and some devote considerable time, effort and

research to finding ways of reducing it.

This is perhaps misplaced in view of the fact that this study found that enhancing unit

values has a far bigger impact on developer profit than reducing costs. Table 4.7

shows the effect of

■ reducing costs by 10% and

■ increasing revenue (GDV) by 10% on each of the case study sites.

Cost reductions had the effect of increasing the residual value per hectare by

between 4 and 8%. Increasing the value of units had a far bigger impact on potential

margins, increasing the residual value per acre by between 14 and 18%.

While this study’s sample is too small to prove that good design necessarily adds value

to residential units, the GDV per hectare on all but one of the exemplar sites was higher

than the conventional comparator. This indicates the possible value of further research

and development to find ways of increasing the gross development value per hectare

rather than concentrating solely on costs. We believe this merits further investigation.


41T H E V A L U E I M P A C T O F H O U S I N G D E S I G N A N D L A Y O U T

Scheme Location Average unit value Unit sizeAverage m2 Average ft2

A Surbiton – exemplar £430,193 130 1,401

B Surbiton – conventional £516,293 151 1,623

C Bishop’s Stortford – exemplar £348,103 150 1,620

D Bishop’s Stortford – conventional £479,646 264 2,843

E Chelmsford – exemplar £238,434 117 1,262

F Chelmsford – conventional £222,024 112 1,207

G Aylesbury – exemplar £155,993 75 807

H Aylesbury – conventional £189,796 90 965

Source: DLE, FPDSavills

Table 4.6: Size of units

A B C D E F G H

Effect of reducing costs -10% 5.05 4.16 6.83 8.09 7.21 6.96 8.40 6.81

Effect of increasing revenue +10% 15.05 14.16 16.83 18.09 17.21 16.96 18.40 16.81

Source: DLE, FPDSavills

Table 4.7: Percentage

increase in residual value on

each site


42

Perhaps of significance here is the way in which developers look at their finances. Few

nowadays look at the absolute profit made on a single site but most would look instead

at return on capital employed in the form of an internal rate of return (IRR). This has an

important impact on the ability of developers to discern how they are making their

money, whether it is the act of building houses or whether it is developing land that is

creating value on a site. The form of analysis used here has begun to suggest that it

may be the latter. Further robust investigation needs to be undertaken on more sites in

order to ascertain this.

Full details and analysis of build, site and marketing costs are contained in Appendix 2

but Table 4.8 summarises them.

When measured on a per standardised hectare basis, the building costs of the

exemplar schemes are usually, but not always, higher than the conventional schemes.

The Surbiton exemplar scheme, A, cost nearly twice as much per hectare as its

matched pair, reflecting the fact that it was built at approximately double the density.

The exemplar cases at Chelmsford, E, and Aylesbury, G, both cost 18% more than the

conventional match. It is interesting to note, however, that the exemplar scheme at

Bishop’s Stortford cost less to build per hectare than the conventional example there,

despite the unit density being higher. This possibly reflects the high development

density (or mass) of the conventional scheme.

4.4 Values4.4.1 Unit values

The notion that higher density decreases value was not supported by this study. In two

of the four cases, the value of the units in the higher density scheme was higher on a

per square metre basis than the comparable lower-density schemes (in one case,

substantially so). Clearly, there is a critical balance to be struck by the developer in

maintaining unit values while increasing density. Some developers managed to achieve


Table 4.8: Total build costsScheme Location Build costs: £000

Per hectare (acre) Per m2 (ft 2)

A Surbiton – exemplar 6,260 (2,533) 1,110 (103)

B Surbiton – conventional 4,467 (1,808) 1,006 (93)

C Bishop’s Stortford – exemplar 3,143 (1,272) 939 (87)

D Bishop’s Stortford – conventional 3,466 (1,403) 812 (75)

E Chelmsford – exemplar 2,761 (1,117) 852 (79)

F Chelmsford – conventional 2,418 (979) 812 (75)

G Aylesbury – exemplar 2,941 (1,190) 950 (88)

H Aylesbury – conventional 2,501 (1,012) 857 (80)

Source: FPDSavills

this while others did not. Striking this balance is the single most important issue in

determining the commercial value of PPG3-compliant schemes for the developer. There

is compelling evidence to advocate that quality design can achieve high value

environments while at the same time increasing density. Build costs are relatively

unimportant in achieving this if value can be maintained.

The value of an individual unit is not a good indication of the value of a scheme to a

developer. Both conventional schemes B and D at Surbiton and Bishop’s Stortford

have average unit values that are considerably higher than the exemplars but the

residual value to the developer per hectare of land is still higher because the unit

densities are higher. Some developers on higher density schemes have achieved a

higher price per square metre than the lower density comparator.



Scheme Location Unit valueAverage Per m2 (ft 2)

A Surbiton – exemplar £430,193 £3,305 (£307)

B Surbiton – conventional £516,293 £3,424 (£318)

C Bishop’s Stortford – exemplar £348,103 £2,313 (£215)

D Bishop’s Stortford – conventional £479,646 £1,816 (£169)

E Chelmsford – exemplar £238,434 £2,033 (£189)

F Chelmsford – conventional £222,024 £1,980 (£184)

G Aylesbury – exemplar £155,993 £2,082 (£193)

H Aylesbury – conventional £189,796 £2,116 (£197)

Source: DL&E, FPDSavills

Table 4.9: Value of units

5.0

Conclusions

C O N C L U S I O N S


46

This work provides an important first step in beginning to assess the value of design

in the commercial world of housing development. It has not provided any clear-cut

results and could never be deemed to have selected and tested a representative

sample. Nevertheless, it has raised important issues about how design is valued

and has indicated how it may make a difference to developers’ profitability.

The first of the broad issues raised by the research is what creates value in the first

place. A model of housing value, developed by FPDSavills’ Research Department

illustrates that there are numerous components of housing value ranging from basic

shelter qualities, through amenity value and luxury value to investment and even

speculative value. Design can clearly impact on the quality of the shelter and the

amenities that a house has, and certainly the extent to which it is deemed to be

luxurious or desirable.

The study highlighted the extent to which one development, particularly those larger

schemes studied, can affect local externalities. There can be a knock-on effect on

the surrounding area affecting not only value but also the sense of place and how a

locality is perceived. It was noticeable that the innovative design elements of one

developer on some parts of a larger development area could influence the value of

the more conventionally designed housing on another part of a site, even though

that housing did not follow the same design principles.

Does this research begin to address the question of how a desirable location may

be created? The endlessly quoted and much loved estate agents’ and TV

presenters’ phrase, ‘location, location, location’ is usually put forward as

emphasising the most important element of housing value. Leaving aside the

macroeconomics, there is a series of factors, including the built form, which creates

a sense of place in a given location. Some of the developers in this study have

helped to show that it is possible to create a sense of place, and hence a new type

of location, on a piece of land by employing certain design principles. These new

places may be more desirable, and therefore more valuable, than their neighbours

and may also be made more desirable and valuable than would otherwise have

been possible had standard development house types and layouts been employed.

There is a strong compulsion to undertake further research using the method

developed in this study to confirm that which is presently only indicative.

Developers and researchers in other locations and other situations have also

noticed that the creation of a distinct place can enhance value. This value

enhancement can spread further than just the site in question. Neighbouring

developers and landowners also benefit. Commercial developments of high quality

on a large scale have pushed up neighbouring land values for both commercial and

residential development; examples include Brindley Place in Birmingham and Canary

Wharf in London. Within this study, the exemplar schemes at Surbiton and

Chelmsford may have achieved this. Other developers on nearby sites may well be

benefiting from the sense of place that these schemes have provided for the

location.

When it comes to value for a developer as opposed to an owner-occupier, it may be

Conclusions


5.0

more appropriate to talk about ‘density, density, density’. The issues involved are

complex but the rules of the game now seem to be to maximise unit values while

increasing the number of units on the site to meet planners’ requirements. It may be

said that this has always been the developers’ game plan but new design

techniques, and indeed changing planning rules, have upped the stakes. Whereas

previously a developer might have aimed for densities of between 25 and 30 per

hectare (10 and 12 per acre), new planning and design technologies have changed

the target to something more like 30–40 per hectare (12–16 per acre). In urban

situations, these densities would be very much higher again. Values will only be

maintained at these densities if design techniques become very much more

sophisticated to maintain the attractiveness to occupiers.

Another question that this research raises is whether design can ever make a really

big value difference in a supply-constrained market. The sites under study were all

in high demand regions of south-eastern England. This region is particularly

characterised by very low levels of new build in relation to increasing demand for

housing. The number of new developments makes a very small impact on overall

supply in those housing markets. In a supply-constrained scenario, it may be that

design has less of an impact than where purchasers’ choices are freer. The question

perhaps is, where the supply of new homes meets demand, whether conventional

designs would sell at a discount or whether enhanced design would sell at a

premium.

The final, and perhaps most important, question is whether this research is really

measuring value to the developer. The margin that the developer makes out of any

scheme is still highly dependent on the price the developer paid for the land. Even

huge residuals would mean nothing if the developer has to pay exceptionally high

prices to secure the land. In cases where competition for land is high (as it is in

much of the South-East), it is likely that any developer with a competitive advantage

(whether it be through design or for any other reason) will have to pass much of that

value on to the landowner. This is because they would need to use enhanced value

as their bidding advantage in order to secure the site against competition in the first

place. Only where developers are part of the longer term land acquisition, ownership

and land development process are they more likely to benefit from any enhanced

value of their design efforts. This is likely to favour more adventurous developers

who can add value at the start of the development process through design and

detailed master-planning and who are prepared to separate their building activity

from this specialised, but potentially lucrative, task.

Ultimately, therefore, it is very likely that if, for any reason, there were to be a

revolution in development values it would only prove advantageous to a developer

while the developer’s competition were unwilling or unable to make use of the

advantage themselves. FPDSavills has already predicted that land values will be

driven up during the next 5–10 years, primarily through development gains of

increased density. If certain other design principles were also discovered to increase

development value, no doubt these would find their way into land values as well.

Not all of the exemplar schemes selected are as dense as PPG3 recommends.





48

While it would appear that good design may have enhanced the schemes’

commercial value, what this study begins to show is that good design becomes

more and more essential to achieve the maximum value as schemes increase in

density, over 30 units per hectare. Standard design and standard layouts will not

work on high density developments. Increases in density must accompany

increased attention to design quality: there are clear commercial rewards when this

course is followed.

There are clearly many questions to be answered and new avenues for this research

to go down. A geographically broader and increased sample would enable more

vigorous testing of hypotheses. It is to be hoped that this study will show how sites,

costs and values can be analysed and that more developers will be willing to offer

their schemes for scrutiny.

Literature review references

Appendix 1.0

A P P E N D I X 1

A P P E N D I X 1

50

Adams, D. (1994) Urban planning and the development process, London, UCL Press

Ball, M. (1983) Housing policy and economic power: the political economy of owner-occupation, London,

Methuen

Bartlett School of Planning, CABE (2001) The value of urban design, London, Thomas Telford

Booth, P. (1982) ‘Housing as a product: design guidance and residential satisfaction in the private sector’, Built

Environment, 8, No. 1, 20–24

Bramley, G., Bartlett, W. and Lambert, C. (1995) Planning, the market and private housebuilding, London, UCL

Press

Carmona, M. (1999) ‘Controlling the design of private sector residential development: an agenda for improving

practice’, Environment and Planning B: Planning and Design, 26, 807–833

Carmona, M. (2001) Housing design quality: through policy, guidance and review, London, E & FN Spon

Cheshire, P. & Sheppard, S. (1997) ‘The welfare economics of land use regulation’, research papers in spatial

and environmental analysis No. 42, Department of Geography, London School of Economics

Cheshire, P. and Sheppard, S. (2000) ‘Building on brown fields: the long term price we pay’, Planning in

London, 33, April/June, 34–36

Civic Trust (1999) Brownfield housing – 12 years on, London, Civic Trust

Colquhoun, I. and Fauset, P.G. (1991) Housing design in practice, Harlow, Longman Scientific and Technical

Conran Roche/Davis Langdon & Everest (1989) Costs of residential development, London, Conran Roche

Davison, I. (1991) Land values: part of the design equation, Housebuilder, March, 38–44

Davison, I. (1990) Good design in housing, London, House Builders Federation/Royal Institute of British

Architects

Department of the Environment/Housing Research Foundation (1976) Design guidance survey: report on a

survey of local authority design guidance for private residential development, London, Department of the

Environment

Department of the Environment (Northern Ireland)/Department for Regional Development (2000) Creating

places: achieving quality in residential environments, Department of the Environment, Northern Ireland

Department of the Environment, Transport and the Regions (1998a) Places, streets and movement: a

companion guide to Design Bulletin 32 — Residential roads and footpaths, London, DETR

Department of the Environment, Transport and the Regions (1998b) Planning and development briefs: a guide

to better practice, London, DETR

Department of the Environment, Transport and the Regions (2000) Planning Policy Guidance Note 3: Housing,

Norwich, The Stationery Office

Department of the Environment, Transport and the Regions/Commission for Architecture and the Built

Environment (2000) By design, Urban design in the planning system: towards better practice, London,

Thomas Telford

Department of Transport, Local Government and the Regions/Commission for Architecture and the Built

Environment (2001) By design: better places to live, London, Thomas Telford

Essex County Council (1973) A design guide for residential areas, Colchester, County Council of Essex

Essex Planning Officers Association (1997) A design guide for residential and mixed use areas, Essex, Essex

Planning Officers Association

Goodchild, G. (1994) ‘Housing design, urban form and sustainable development’, Town Planning Review,

65 (2) 143–158

Goodchild, B. (1997) Housing and the urban environment: a guide to house design, renewal and urban

planning, Oxford, Blackwell Science

a1.0


Literature reviewreferences


Hall, A. (1990) Generation of objectives for design control, Chelmsford, Anglia College Enterprises Ltd

Hanson, J. (1999) Decoding homes and houses, Cambridge, Cambridge University Press

Hooper, A. & Nicol, C. (1999) ‘The design and planning of residential development: standard house types in

the speculative housebuilding industry’, Environment and Planning B: Planning and Design, 26, 793–805

Hooper, A. & Nicol, C. (2000) ‘Design practice and volume production in speculative housebuilding’,

Construction Management and Economics, 18, 295–310

Karn, V. (1992) ‘British housing standards: time for a new approach?’ in Housing design — time for a new

approach, London, Housing Centre Trust

Karn, V. and Sheridan, L. (1994) New homes in the 1990s: a study of design, space and amenity in housing

association and private sector production, University of Manchester/Joseph Rowntree Foundation

Leopold, E. and Bishop, D. (1983a) ‘Design philosophy and practice in speculative housebuilding: Part 1’,


Leopold, E. and Bishop, D. (1983b) ‘Design philosophy and practice in speculative housebuilding: Part 2’,


Llewelyn-Davies (1994) Providing more homes in urban areas, University of Bristol, SAUS Publications

Llewelyn-Davies (1998) Sustainable residential quality: new approaches to urban living, London, London

Advisory Planning Committee

Llewelyn-Davies (1999) Sustainable residential quality II: approach and method statement, London, London

Advisory Planning Committee

Llewelyn-Davies (2000a) Sustainable residential quality: exploring the housing potential of large sites, London,

London Advisory Planning Committee

Llewelyn-Davies (2000b) Urban design compendium, London, English Partnerships/The Housing Corporation

Nicol, C. & Hooper, A. (1999) ‘Contemporary change and the housebuilding industry: concentration and

standardisation in production’, Housing Studies, 14, No. 1, 57–76

Parfect, M. and Power, G. (1997) Planning for urban quality: urban design in towns and cities, London,

Routledge

Planning Officers Society, Housebuilders Federation and DETR (1998) Housing layouts — lifting the quality,

London, HBF

Popular Housing Forum (now Popular Housing Group) (1998) Kerb appeal: the external appearance and site

layout of new houses, Winchester, Popular Housing Group

Punter, J. and Carmona, M. (1997) The design dimension of planning: theory, content and best practice for

design policies, London, E & FN Spon

RICS (1996) Quality of urban design: a study of the involvement of private property decision-makers in urban

design, London, Royal Institution of Chartered Surveyors

Rudlin, D. & Falk, N. (1995) 21st century homes: building to last, York, Joseph Rowntree Foundation

Rudlin, D. & Falk, N. (1999) Building the 21st century home: the sustainable urban neighbourhood, Oxford,

Architectural Press

Sim, D. (1993) British housing design, Harlow, Longman

Syms, P. (2001) Releasing brownfields, London, RICS Foundation

Syms, P. and Knight, P. (2000) Building homes on used land, London, RICS Books

Urban Task Force (1999) Towards an urban renaissance, London, E & FN Spon

Verhage, R. and Needham, B. (1997) ‘Negotiating about the residential environment: it is not only money that

matters’, Urban Studies, 34, No. 12, 2053–2068

Worpole, K. (1999) The value of architecture — design, economy and the architectural imagination, RIBA

Future Studies Series, London, RIBA

L I T E R A T U R E R E V I E W R E F E R E N C E S

Cost analysis by Davis Langdon & Everest

Appendix 2.0

A P P E N D I X 2

A P P E N D I X 2

54

A2.1 Introduction This section of the report discusses the objectives and scope of the analysis of

housing development costs. The methodologies adopted for data collection,

analysis and presentation are outlined. Issues associated with the analysis of the

data and the results, which could have an influence on the overall outcome of the

study, are also identified.

A2.2 Objective The objective of this cost assessment is to establish benchmark development costs

for each scheme using formats that facilitate direct comparison. The adjustment and

analysis of the development cost data provides the basis on which an assessment

of net development value and overall land value can be calculated. Issues that need

to be addressed in preparing this analysis include:

■ the establishment of a common data structure

■ ensuring that the full scope of relevant development costs are captured for

each project

■ adjusting costs for time and location to enable direct comparisons to be made

between matched pairs of developments

A2.3 Data collection Construction and development cost data has been collected directly from a group

of contributing housebuilding companies. Interviews were held with commercial

managers, during which detailed cost and project information was collected. These

meetings gave Davis Langdon & Everest’s (DL&E’s) researchers the opportunity to

collect additional information related to market conditions and other variables that

could potentially affect the overall development cost assessment.

No standard cost breakdown structure is available for use on private housing

projects. The information collected for this project has, as a result, been provided in

a number of formats, often related to the costs of individual trades or tranches of

building materials. The researchers have re-categorised this information to permit

direct comparison between projects.

The information provided by the developers is potentially very commercially

sensitive. The team’s awareness of this consideration has influenced the extent to

which detailed analysis of the development costs information can be undertaken as

follows:

■ costs are reported on a scheme-wide basis; specific costs for individual

dwelling types are not published

■ costs are reported using indirect units of comparison (cost per square foot of

gross internal floor area or cost per acre)


a2.0 Assessment ofdevelopment costs

In two locations, it has not been possible to obtain detailed project and cost

information for a pair of comparable projects. In these circumstances, models have

been developed based on design information available in the public domain and

comparable cost information derived from the rest of the sample. The issues

associated with the preparation of the models are discussed further in section A2.5

below.

A2.4 Scope and structure of project cost data In the preparatory analysis, cost data collected from each developer has been

adjusted to include only cost centres that relate to the development cost of specific

phases of development. For example, in order to facilitate comparison between

schemes, costs of primary infrastructure and site development costs have been

excluded. The following cost centres are included in the analysis:

■ construction cost — building only

■ construction costs — external works and services constructed as part of the

development

■ phase featured in the sample

■ construction costs — preliminaries (contractor’s on-site management

resources)

■ internal development costs including:

— designers’ fees, fees associated with planning and NHBC inspections, etc

— internal marketing costs including show homes, sales staff costs and

advertising, etc

Items excluded from the analysis include:

■ remediation works

■ site development costs beyond the boundary of the site/phase — this

includes all section 106 and section 278 works associated with the

development

■ for larger schemes, where the sample comprises one or more phases of a

larger development, this means that the costs of site preparation and strategic

infrastructure works required to prepare the site for development are also

excluded, as are communal facilities such as parks and shelter belts which

benefit other phases of the scheme

■ for smaller, or self-contained schemes, all infrastructure within the site

boundary is included, on the basis that the extent of works required will be

equivalent to that of a single phase on a larger housing development

■ finance costs

■ costs of head office overheads and profit

The exclusion of these items from the scope of the analysis does not affect the

validity of the outcome. By removing site- and contractor-specific variables, such as

C O S T A N A L Y S I S B Y D A V I S L A N G D O N & E V E R E S T


A P P E N D I X 2

56

those related to site conditions, infrastructure and planning requirements, and costs

of development, capital costs based on common criteria can be directly compared.

While this represents a simplification of the total cost equation, use of this

assumption ensures that the analysis identifies relationships between scheme

design, density building cost and development value. In particular, the exclusion of

planning gain costs, etc removes a significant source of site-specific variation from

the data set.

A2.5 Analysis of the dataA2.5.1 Normalisation

The primary objective of the analysis is to provide consistent, comparable

development cost information for input into the overall assessment of development

value. The data is also presented in a range of formats to facilitate the comparison

of schemes in the sample.

Adjustments required to ensure that the cost information is comparable are as

follows:

■ Inflation — development costs have been brought to a common datum

(March 2002) using DL&E’s published tender price index. The adjustments

used were verified using information provided by feedback from the

developers’ commercial managers

■ Location — cost differentials between locations have been updated to levels

current in March 2002 using location factors published by the Building Cost

Information Service of the RICS

■ Common units of comparison — gross internal floor area and overall site

area have been used as the units of comparison

A2.5.2 Development of cost models

It has not been possible to obtain cost information for two schemes in the sample.

As an alternative, cost models have been prepared to provide the basis for the

comparison. The key features of the cost models are:

■ the size, layout and mix of dwellings is based on a specific, directly

comparable scheme

■ costs are derived from appropriate schemes elsewhere in the sample

■ cost models reflect differences in quantities related to key cost drivers for each

house type

■ models were developed for each house type in a scheme

The models were developed from drawn information and other marketing material

available in the public domain. While the research team acknowledges that the

models are generic and cannot account for all minor variations in detailed design,

these variations only have a limited cost impact. Given the circumstances of the


study, the models represent the most effective method of obtaining cost information

with an acceptable level of accuracy.

In compiling the models, DL&E have also taken into account differences in external

works content — particularly road and sewer layouts. The assessment of other

development costs (fees, marketing costs, etc) is based on a review of the

proportion of costs recorded on the other projects in the sample.

A2.5.3 Presentation of the data

The results of the analysis are published in three formats.

■ Total cost per square foot of gross internal floor area. This analysis

provides a direct comparison of development cost based on the average

building cost for each scheme. A number of variables will affect these costs,

including some which are site-specific rather than related to the design of the

house type such as:

— extent of substructure works related to local ground conditions

— extent of external works, which are affected by both the extent and

quality of works in the public domain, and by the development density

of the scheme

The inclusion of these costs makes it difficult to identify differences in the

construction costs related to specification. These can be identified more

clearly in an analysis of superstructure costs

■ Superstructure only costs per square foot of gross internal floor area.

This analysis eliminates the variables of substructure and external works and

enables a direct comparison to be made between the costs associated with

the build quality of the particular dwelling mix

This comparison excludes the key element of the external works, which clearly

contribute to value through the provision of a high quality public realm.

External works need to be assessed using a different unit of comparison —

site area — to provide meaningful results

■ Total costs per acre. This analysis enables the effect of the key variable of

development density to be accounted for. External works costs can be

compared directly on the basis of the site area rather than gross internal floor

area (gifa)

A2.6 ResultsTable A2.1 sets out the total development costs for the eight schemes in the

sample. The projects are organised as matched pairs in accordance with the study

methodology outlined elsewhere in this report. The costs of each matched pair of

schemes have been brought to a common datum related to their location. Table

A2.2 uses a similar format to present superstructure-only costs.



A P P E N D I X 2

58

Market conditions and price levels in the locations of schemes C to H are similar

and these projects can be compared more or less directly. However, market

conditions and other factors affecting costs of projects in south-west London mean

that the costs of schemes A and B are subject to a 5% premium compared to

others in the sample.


Table A2.1: Total summary

construction costs

Scheme A Scheme B Scheme C Scheme DSouth-west London East Hertfordshire

Site area (acres) 20.51 12.18 5.96 3.65Development area (ft2) 442,786 188,266 93,947 68,243

Development density 0.496 0.355 0.362 0.429

Development cost £/ft2 % £/ft2 % £/ft2 % £/ft2 %

Foundations/ground slabs 6.28 6.09 3.78 4.05 5.99 6.86 3.76 4.99Masonry 17.52 17.00 11.32 12.12 9.72 11.15 10.49 13.90Carpentry/windows/doors 9.36 9.08 9.45 10.12 9.83 11.27 12.10 16.04Roof coverings 2.60 2.53 3.82 4.09 3.45 3.95 2.62 3.48Joinery, fixtures and fittings 3.27 3.17 2.13 2.28 8.04 9.22 1.86 2.46Finishes 13.52 13.11 9.88 10.58 8.38 9.61 7.52 9.96Kitchen installation 4.88 4.74 3.57 3.82 4.45 5.11 3.26 4.32Services 11.60 11.26 9.91 10.60 6.52 7.48 8.73 11.57Preliminaries 13.61 13.20 19.56 20.93 8.54 9.79 7.64 10.13Site preparation 0.92 0.89 0.49 0.52 0.00 0.00 0.36 0.47External works 6.79 6.59 7.74 8.29 7.67 8.80 4.78 6.33Drainage and below ground services 3.03 2.94 4.56 4.88 3.88 4.45 3.63 4.81Internal development costs 9.69 9.40 7.21 7.71 10.72 12.29 8.71 11.54(fees, planning legal, NHBC, inspections, customer care)

Total Build Cost 103.08 100.00 93.42 100.00 87.20 100.00 75.44 100.00

Scheme E Scheme F Scheme G Scheme HEssex Buckinghamshire




Foundations/ground slabs 10.04 12.68 5.54 7.34 7.18 8.14 5.59 7.02Masonry 11.52 14.56 10.09 13.36 9.58 10.85 10.49 13.17Carpentry/windows/doors 10.42 13.16 11.59 15.35 10.82 12.25 12.61 15.84Roof coverings 4.67 5.90 3.11 4.12 3.38 3.82 2.87 3.60Joinery, fixtures and fittings 2.24 2.83 1.42 1.89 1.70 1.93 1.45 1.82Finishes 7.22 9.12 5.75 7.62 6.43 7.28 5.86 7.36Kitchen installation 4.50 5.69 1.79 2.37 2.55 2.89 2.11 2.65Services 7.63 9.65 8.56 11.34 10.10 11.44 8.88 11.15Preliminaries 7.96 10.06 7.16 9.49 8.34 9.45 8.91 11.20Site preparation 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00External works 3.82 4.83 7.19 9.52 11.68 13.23 8.56 10.75Drainage and below ground services 2.89 3.65 4.60 6.09 6.26 7.09 3.13 3.93Internal development costs 6.23 7.88 8.68 11.50 10.26 11.62 9.16 11.50(fees, planning legal, NHBC, inspections, customer care)

Total Build Cost 79.13 100.00 75.48 100.00 88.27 100.00 79.62 100.00

■ The table details average building costs for all dwellings on eight housing schemes in south-east England. Costs are given on the basis of cost per square foot of grossinternal floor area

■ Costs are stated at March 2002 price levels, and reflect the costs associated with the locality of each pair of projects. The price level reflects the completion of the housing asa single phase for sale in March 2002

■ The costs in this table include all project specific development costs associated with each scheme, including all external works required to develop the plot, professional feesand sales and marketing expenses

■ The analysis does not include costs of strategic site infrastructure or section 106 and 278 works. The analysis also excludes funding costs and overhead contributions

The results of the analysis summarised in Tables A2.1 and A2.2 show that the four

urban design exemplar schemes are consistently more expensive to construct than

their conventional equivalents. The only exceptions are the superstructure building

only costs associated with schemes G and H in Essex. The overall cost differential

is set out in Tables A2.3 and A2.4.



Table A2.2: Summary

superstructure construction

costs

Scheme A Scheme B Scheme C Scheme DSouth-west London East Hertfordshire




Masonry 17.52 27.92 11.32 22.61 9.72 19.30 10.49 22.52Carpentry/windows/doors 9.36 14.92 9.45 18.87 9.83 19.50 12.10 25.99Roof coverings 2.60 4.15 3.82 7.62 3.45 6.84 2.62 5.63Joinery, fixtures and fittings 3.27 5.21 2.13 4.26 8.04 15.96 1.86 3.98Finishes 13.52 21.54 9.88 19.74 8.38 16.63 7.52 16.14Kitchen installation 4.88 7.78 3.57 7.13 4.45 8.83 3.26 7.00Services 11.60 18.49 9.91 19.78 6.52 12.94 8.73 18.74

Superstructure, fittings and finishes costs 62.76 100.00 50.09 100.00 50.40 100.00 46.57 100.00

Scheme E Scheme F Scheme G Scheme HEssex Buckinghamshire




Masonry 11.52 23.90 10.09 23.84 9.58 21.50 10.49 23.69Carpentry/windows/doors 10.42 21.61 11.59 27.39 10.82 24.28 12.61 28.49Roof coverings 4.67 9.69 3.11 7.35 3.38 7.58 2.87 6.47Joinery, fixtures and fittings 2.24 4.65 1.42 3.37 1.70 3.82 1.45 3.28Finishes 7.22 14.98 5.75 13.60 6.43 14.43 5.86 13.24Kitchen installation 4.50 9.34 1.79 4.22 2.55 5.73 2.11 4.77Services 7.63 15.84 8.56 20.24 10.10 22.67 8.88 20.06

Superstructure, fittings and finishes costs 48.19 100.00 42.31 100.00 44.56 100.00 44.27 100.00

■ The table details average building costs for all dwellings on eight housing schemes in south-east England. Costs are given on the basis of cost per square foot of grossinternal floor area

■ Costs are stated at March 2002 price levels, and reflect the costs associated with the locality of each pair of projects. The price level reflects the completion of the housing asa single phase for sale in March 2002

■ The costs in this table include costs associated with the superstructure, finishes and services of the houses in each scheme only. All below ground works, preliminaries andexternal works required to develop the plot, professional fees and sales and marketing expenses are excluded. This analysis makes it easier to identify differences inspecification between each scheme. (Refer to Table 1 for the total costs)

■ The analysis does not include costs of strategic site infrastructure or section 106 and 278 works. The analysis also excludes funding costs and overhead contributions

A P P E N D I X 2

60

The further analysis of the differentials between the costs of the matched pairs is set

out in section A2.6.2, Analysis of matched pairs.

A2.6.1 Sources of variation in building costs

The comparison of the construction costs of low-rise residential schemes is

relatively straightforward, as there is a limited number of variables that can affect the

overall level of cost. The principal variables are detailed below.

■ Site conditions and foundations design. The design and costs of

foundations are generally influenced more by ground conditions than by the

specific design, layout and massing of dwellings. Poor ground conditions can

add significantly to construction costs. However, substructure costs are easily

identified and can be isolated from the comparison if required, as they have

been from Table A2.2

■ Efficiency of layout. The efficiency of house layout effectively determines

how much building superstructure is required to enclose each square foot of

accommodation. More efficiently designed dwellings will, all other factors

being equal, have relatively low external wall and roof costs. Although the

differences can be marginal, the cost effect can be more significant due to the

direct effect on the costs of masonry, carpentry and roof coverings, which

account for a high proportion of overall construction costs. Other factors

which affect building efficiency include the facts that:

— larger dwellings tend to be slightly more efficient than smaller versions of

the same layout

— standard developer homes based on a square plan are particularly efficient

— the introduction of rooms in the roof increases the overall efficiency of

house layout and design


Schemes A & B Schemes C & D Schemes E & F Schemes G & H

Exemplar scheme (£/ft2) 62.76 50.40 48.19 44.56

Conventional scheme (£/ft2) 50.09 46.57 42.31 44.27

Differential (%) 80.0 92.0 87.8 99.0


Table A2.4: Differential in total

construction costs —

superstructure costs only

Table A2.3: Differential in total

construction costs — total

development cost

Schemes A & B Schemes C & D Schemes E & F Schemes G & H

Exemplar scheme (£/ft2) 103.08 87.20 79.13 88.27

Conventional scheme (£/ft2) 93.42 75.44 75.48 79.62

Differential (%) 90.6 86.5 95.4 90.2


■ Specification of materials, fixtures and fittings. The cost breakdown

designed for this study enables significant variations in the cost of elements to

be identified. Some of these differences will be due to specification variations,

which may also affect development value. For the purposes of this study, it is

not necessary to quantify differences between the detailed design of matched

pairs, however, this analysis provides the opportunity to identify where

developers of high cost/higher value properties are spending in response to

consumer demand

■ Kitchens. The square foot cost of kitchens and white goods is generally

higher in smaller dwellings

■ External works. The extent and mix of external treatments, and potential

range of materials and specifications mean that external works also need to

be clearly identified and isolated to enable direct comparison of house

construction costs. External work costs, analysed on the basis of cost per

square foot, are also affected by the density of development, with two factors

in particular combining to affect the overall rate:

— an increasing proportion of hard landscaping as schemes increase in

density, resulting in higher costs

— the reduction in square foot costs of external works generally on

schemes with a high proportion of gross floor area to overall site area

These factors are illustrated in Table A2.5 which sets out the densities and costs

(per square foot and per acre) for external works on each scheme.

Table A2.5 illustrates the range of costs recorded for external works, and the changing

relationships between these costs related to the unit of comparison adopted.

A2.6.2 Analysis of matched pairs

This section of the report sets out a more detailed discussion of the costs of the

projects, based on the data sets detailed in Tables A2.1 and A2.2.

1. Surbiton, south-west London

Schemes A & B are the only developments located in an urban setting. Situated in

south-west London, these are the two most costly schemes in the sample, even



A B C D E F G H

Density 0.496 0.355 0.362 0.429 0.320 0.324 0.301 0.306

Rate (£/ft2) 6.79 7.74 7.67 4.78 3.82 7.19 11.68 8.56

Rate (£/acre) 146,614 119,662 120,956 89,308 53,224 101,288 152,858 113,948


Table A2.5: Summary costs of

external works

A P P E N D I X 2

62

when relative cost levels are taken into account. Schemes A & B are the only

projects that feature apartments as part of the project. Scheme A is also the

densest in the sample. The impact of density can be seen in the relatively low

proportion of expenditure on external works, for a scheme that has the most

extensive landscaping investment. (Cost of landscaping per acre is £146,614

compared to a normalised sample average of £108,916 per acre.)

Relatively high levels of capital expenditure on scheme A are focused on

substructure (associated with undercroft car parking), masonry, joinery and finishes,

reflecting a high level of specification throughout the scheme, which is likely to drive

up overall development value. Scheme B costs are fairly similar to those of projects

outside London. A very high proportion of the costs of this scheme is associated

with preliminaries — which potentially could include items of unstated direct labour

or the sourcing of materials. Table A2.2 sets out superstructure building works only

costs for the sample of eight projects. Adjusted for location (£47.60 per ft2 ), the

costs of scheme B are very similar to those found on projects C to H. Notably, high

levels of expenditure on this scheme are directed at internal finishes.

2. Bishop’s Stortford, East Hertfordshire

Schemes C & D are both high quality developments located on estates on a

rural/urban fringe. Both incorporate elements of PPG3, with dwellings in scheme C

being built up to the street-line, and scheme D featuring houses with rooms in the

roof. Both schemes are designed to a relatively high level of density when measured

on the basis of the ratio of floor area to site area. However, the dwellings on both

schemes are large and neither scheme meets the density targets set in PPG3.

Scheme C features houses strongly influenced by the vernacular design and also

incorporates extensive hard landscaping. The cost breakdown in Table A2.1 shows

that the costs of scheme C are relatively high — demonstrating that the

construction of non-standard housing does involve a cost premium. From the

analysis in Table A2.2, there is evidence of an across-the-board approach to

investment in quality, with higher than average expenditure on joinery, fixtures and

fittings, and finishes.

Scheme D features large four- and five-bed houses. Overall costs are relatively low,

partly due to the large size of the dwellings and also due to the influence of rooms

in the roof on the efficiency of the design and its use of building materials. Table

A2.2 shows that there are particularly low levels of expenditure on roof coverings,

joinery and kitchens. The low cost of roofing and kitchen fittings is often a result of

these costs being amortised over an especially large floor area. The low cost of

joinery and fittings may be as a result of some items being included in the carpentry

section. Overall, the breakdown shows slightly higher expenditure on masonry and

services — areas that directly affect building quality and occupant comfort

respectively. Low external works costs can be associated with the simplicity of the

cul-de-sac design concept and the high density (in terms of relative area, not

number of dwellings) of the scheme.


3. Chelmsford, Essex

Schemes E & F are phases of development of a large site on a rural/urban fringe.

The design of the layout of both developments is strongly influenced by principles

laid out in the Essex Design Guide. The developments feature a mix of three- and

four-bed dwellings. Only limited information has been provided for scheme E by the

developer. The scheme features a wide variety of units in complex configurations.

While overall scheme development costs, at £79/ft2, are relatively low, the

superstructure building costs, at £48/ft2, are among the highest in the sample. High

levels of expenditure are recorded on substructures, masonry,

carpentry/windows/doors and the kitchen installation. Table A2.2 highlights this

pattern of expenditure.

The construction costs for scheme F are based on the output of a generic model. It

should be noted that the costs of substructure are also based on a generic solution

as no site details or site-specific costs are available. If the substructure costs of

schemes E & F were identical, then the overall construction cost of scheme F would

be just over £80/ft2, i.e. slightly higher than scheme E. This addition to total

construction cost would have a significant impact on development value and funds

available for land purchase. The external works costs for scheme F are the higher of

the pair, related to costs of hard paving to shared driveways and some shared open

space. In both schemes, extensive provision of shared open space, either in

parks/greens, or in heavily landscaped pedestrian avenues, are excluded from the

analysis, as these facilities are provided as part of the overall site infrastructure

rather than as a site-specific amenity.

In the model, while absolute costs for scheme F, detailed in Table A2.2, are lower

than for scheme E, the allocation between cost centres is consistent, with the

exception of expenditure on the kitchen and services.

4. Aylesbury, Buckinghamshire

The costs of scheme H are based on the output of a model, with costs based on

schemes within the sample. Table A2.2 shows that there is little difference between

the costs of superstructure components of the model scheme and scheme G. The

primary source of differential is in the external works and below ground services,

where costs on scheme G are the highest in the sample, featuring substantial levels

of investment in hard landscape, boundary walls, etc. The houses in scheme G are

standard dwelling types with enhancements to masonry and roof specification.



Appendix 3.0

A P P E N D I X 3

Characteristics of each site


A P P E N D I X 3

66

Site A

Land use by size of footprint

Land use As a percentage of total size

Public open space 20.3%

Peripheral and feeder roads, access roads and cul-de-sacs 13.4%

Public driveways, parking courtyards and shared space 1.6%

Private driveways 9.0%

Garages 0.4%

Private gardens 37.0%

Houses and flats 18.3%

Location Surbiton, outer south-west London

Type Exemplar. Real, existing site. Stand-alone scheme

Developer St James Homes

Total size of site: 9.1 hectares (22.6 acres)(all the land the developer would have bought, including feeder and access roads but excluding major bypasses and distributor roads)

Standardised size of site: 7.3 hectares (18.0 acres)(as above but excluding public open space such as parks, greens, verges and recreation ground)

Developer’s stated size of site: 8.3 hectares (20.5 acres)(where applicable)


C H A R A C T E R I S T I C S O F E A C H S I T E

67

Number of units

Type All: 5+ beds 4 beds 3 beds 2 beds 1 bed

Detached 63 38 25 — — —

Semi-detached — — — — — —

Terraced 175 — 175 — — —

Apartments 78 — — — 36 42

TOTAL 316 38 200 — 36 42

Densities On the basis of

Whole site area Standardised site area

Dwellings (units) per hectare (acre) 35 (14) 43 (18)

Habitable rooms per hectare (acre) 203 (82) 255 (103)

Bed spaces per hectare (acre) 190 (77) 239 (97)

Plot ratio (coverage) 0.564

(Gross internal area built as proportion of site area)

Number of habitable rooms 1,859

Number of bed spaces 1,740

Area of all residential buildings (gross internal area) 41,136 m2 (442,786 ft2)

Site B




Peripheral and feeder roads 15.1%

Access roads and cul-de-sacs 1.3%



Garages 1.8%



Location Surbiton, outer south-west London

Type Conventional. Real, existing site. Large phase

on larger scheme

Developer Fairview

Total size of site: 4.4 hectares (10.9 acres)

(all the land the developer would have bought, including

feeder and access roads but excluding major bypasses and

distributor roads)

Standardised size of site: 3.9 hectares (9.7 acres)

(as above but excluding public open space such as parks,

greens, verges and recreation ground)

Developer’s stated size of site: 4.9 hectares (12.2 acres)

(where applicable)

A P P E N D I X 3

68 T H E V A L U E O F H O U S I N G D E S I G N A N D L A Y O U T

Number of units


Detached 44 26 18 — — —

Semi-detached 18 6 12 — — —

Terraced 42 2 40 — — —

Apartments 12 — — — 6 6

TOTAL 116 34 70 — 6 6








Number of habitable rooms 806

Number of bed spaces 751





A P P E N D I X 3

Site C





Access roads and cul-de-sacs, public driveways, parking

courtyards and shared space 15.7%


Garages 5.7%



Location Bishop’s Stortford, East Hertfordshire

Type Exemplar. Real, existing site. Large, late phase onlarger scheme

Developer Countryside Properties






Number of units


Detached 18 4 13 1 — —

Semi-detached 31 — 14 15 2 —

Terraced 9 — — 9 — —

Apartments 0 — — — — —

TOTAL 58 4 27 25 2 —












Site D








Garages 4.0%



Location Bishop’s Stortford, East Hertfordshire

Type Conventional. Real, existing site. Stand-alone

scheme

Developer Anonymous




distributor roads)





(where applicable)

A P P E N D I X 3


Number of units


Detached 24 24 — — — —


Terraced — — — — — —

Apartments — — — — — —

TOTAL 24 24 — — — —













A P P E N D I X 3

74

Site E






Private driveways, garages and private gardens 51.6%


Location Chelmsford, Essex

Type Exemplar. Real, existing site. Large phase on a larger scheme

Developer Bryant Homes






75

Number of units


Detached — — — — — —




TOTAL 96 6 35 22 33 —












Site F




Peripheral and feeder roads, access roads and cul-de-sacs,

public driveways, parking courtyards and shared space 17.5%

Private driveways, garages and private gardens 57%


Location Chelmsford, Essex

Type Conventional. Hypothetical site. Large phase on

a larger scheme

Developer N/A




distributor roads)




Developer’s stated size of site: N/A

(where applicable)

A P P E N D I X 3


Number of units






TOTAL 101 23 27 51 — —













A P P E N D I X 3

Site G







Garages 4.4%



Location Aylesbury, Buckinghamshire

Type Exemplar. Real, existing site. Small, late phase on a larger scheme

Developer Bryant Homes






Number of units






TOTAL 74 — 4 39 28 3












Site H




Peripheral and feeder roads, 14.7%




Garages 4.0%



Location Aylesbury, Buckinghamshire

Type Conventional. Hypothetical. Stand-alone scheme

Developer N/A




distributor roads)





(where applicable)

A P P E N D I X 3


Number of units


Detached 39 — 31 8 — —

Semi-detached 24 — — 17 7 —

Terraced 13 — — — 13 —


TOTAL 76 — 31 25 20 —












Appendix 4.0

A P P E N D I X 4

Step-by-step methodology

A P P E N D I X 4

84

1 Four locations in the south-east of England were selected.

2 An exemplar scheme in each area was then selected on the basis that eachrepresented best practice and/or innovative design and layout and largely followed theurban design principles of PPG3.

3 A conventional housing scheme was then selected near to each of the exemplarschemes.

4 The build costs for both schemes in each area were measured. These costs excludedmajor remedial and infrastructure works and the costs of any section 106 agreements,including social housing (as these differ between sites). They did include landscapingcosts and the costs of site roads, where they were provided by the developer. (A smallflaw in this methodology is that it does not include the cost of site roads and layoutswhere they were provided by another developer. Ideally, they should be taken intoaccount.)

5 Where costs were not available for the standard schemes, Davis Langdon & Everestconstructed a hypothetical model of the scheme and its costs.

6 In each location, three local agents gave an open-market value for all the units on bothof the schemes as at March 2002 as if they were both built on the exemplar land.

7 Where possible, actual values that were achieved on the schemes during March 2002were obtained.

8 These real revenues were checked against the values given by the three local agents. Ifany valuer gave values which differed from the actual revenues by more than 20%, allthe values given by that valuer were removed and not used. (In practice, this happenedin just one case.)

9 The remaining values for each unit were averaged and then all unit average valueswere added together to give the total, gross development value (GDV).

10 The total build costs were subtracted from the total revenue to create a residual. (Thisresidual represented the land value, central overheads and developer profit/margin —not just land value.)

11 The measurement of land was standardised on each site to:

— exclude major roads, bypasses, etc

— include all site roads, including minor peripheral roads

— measure roads on the periphery of the site which were shared with adjacentdevelopments to the centre only

— exclude public open land regardless of whether the developer developed andlandscaped it or not

— exclude land used for any section 106 agreements (such as land used forschools, community facilities and social housing, for example)

12 The residual value arrived at in 10 above was then divided by the number ofhectares measured on the standardised basis derived in stage 11 above.

13 The resulting measure of the residual per hectare (or acre) then provided a measureof landowner/developer value that could be directly compared with other schemes inthe same area and in other areas.

14 The exemplar schemes were compared with the conventional schemes in eachlocation.

15 The differences between each scheme were analysed.

16 The possible impact of these differences on the residual value was examined.


a4.0 Step-by-stepmethodology

Appendix 5.0

A P P E N D I X 5

Illustrative example of the residual land calculation

A P P E N D I X 5

86

Site X would appear to have a greater commercial value, developed in this way,

than site Y.

The commercial value to the developer could be compared between sites in

different locations by subtracting a standard land value figure for each location from

the residual. The remaining figure would represent the profit and/or extra land

buying power (which might enable the developer to secure the site against

competition from other developers at the purchase stage) that is available to the

developer.


a5.0 Illustrative example ofthe residual landcalculation

Site X Site Y

A Gross development value (GDV) £70,000,000 £30,000,000

B Land area (HSSM) 3.5 2

C Total build costs £25,000,000 £9,000,000

D Residual value (A - C) £45,000,000 £21,000,000

E SDV per HSSM (D/B) £12,857,000 £10,500,000

F Percentage difference compared to other site + 22.4% – 18.3%

Terms used:

GDV (gross development value) Total of all average units values given by agents/valuers as ifthey were all built on the same site

Land area Area of land using the standardised measure of site area(outlined in Appendix 4)

HSSM Hectares of standardised site measurement

Total build costs Total cost of developing the private housing units as defined inAppendix 2

Residual value Sum available to the developer to cover land costs, centraloverheads and profit

Standardised development Residual value as defined abovevalue (SDV)

Acknowledgements

A C K N O W L E D G E M E N T S

A C K N O W L E D G E M E N T S

88

AcknowledgementsThanks are due to Davis Langdon & Everest for their contribution to the cost analysis in

this study. Also, to Professor Alan Hooper at Cardiff University for the Literature Review

and the helpful comments and encouragement he gave at the outset of the study.

This research would not have been possible without the help and cooperation of the

developers of both the exemplar and conventional schemes in the case studies. Our

particular thanks are due to Bryant Homes, Countryside Homes, Fairview, St James

Homes and another developer, who prefers to remain anonymous, for their time,

effort and input into this study.

Thanks are also due to the valuers at various offices of estate agents: Bradford and

Bingley, Brown and Merry, FPDSavills, Nicholas Pallet, Peter Anstee & Co and Your

Move.

Finally, the authors would like to thank CABE for their considerable time and effort in

steering this project over the past year or so and to thank them, ODPM, Design for

Homes and the House Builders Federation for sponsoring the study.

Steering GroupDavid Birkbeck Design for Homes

John Miles Ove Arup Partnership

John Weir Haute Future

Christina Golton ODPM

Stephen Mullin ODPM

Simon Pinnegar ODPM

Kate Anderton CABE

Ben Castell CABE

Jon Rouse CABE

Photography creditsPhotographs on front cover and pages 7, 11, 15, 25, 35, 45, 49, 53, 65, 83, 85 and

87 courtesy of Mark Ellis

Photograph of Site A, page 66, courtesy of PRP + Daniel Thistlethwaite


By 2016 an additional three million homes will be required in the UK. Yet last

year an estimated 160,000 new homes were built, far short of the

Government’s target of 225,000.

This report, produced by FPDSavills Research for CABE and ODPM in

association with Design for Homes, seeks to demonstrate a way of measuring


pilots a methodology that developers, funders, planners and housebuilders can

apply to compare housing sites and schemes.

The report is especially timely given pressure from the Government who rightly

insist that new housing development, especially in the south east of England,

should be aiming for densities of over 30 units per hectare. Investing in the

design of these new homes is imperative if we are to deliver quality


ThomasTelford




the value of housing design and layout - the national...

Documents